@@ -48,6 +49,55 @@ nested deeper). The daily cron job `1a8e90ea2271` is `active` with `last_status:
- Page includes "auto-regenerated, do not hand-edit" warning
- Page includes "auto-regenerated, do not hand-edit" warning
- Same script pattern is reusable for other inventory pages (cron jobs, services, agents)
- Same script pattern is reusable for other inventory pages (cron jobs, services, agents)
## Reopened 2026-10-23 — the count includes archived skills, and two generators disagree
The 2026-10-21 all-clear above was premature. Two defects were confirmed on the
night of 2026-10-22/23:
**1. The stocktake ran six times long and committed nothing.** Cron job
`1a8e90ea2271` ran **03:00:28 → 04:38:19 AWST on 2026-10-22 (97 m 58 s)**, against
**16 m 01 s** on 2026-10-21, and finished `completed` with delivery `suppressed` —
but `git status` in the wiki repo still shows the run's regenerated
`systems/skills-index.md` (mtime 10-22 03:00) **uncommitted**, and none of the fixes
the run had diagnosed were applied. A `completed` row is therefore not evidence that
the stocktake did its job.
**2. Two scripts generate the same page with different shapes.**
| Generator | Skill scan | 2026-10-22 output |
|---|---|---|
| `/home/hermes/workspace/agent-estate-wiki/wiki-maintain.sh` (the one the cron job executes) | `find … -name SKILL.md` — does **not** prune dot-directories | **347** skills, with a `## .Archive (41)` section |
- **Current as of 2026-10-22 00:1x AWST (`jobs.json.updated_at 2026-10-22T00:11:31+08:00`) — the fleet is clean except for one failure, and it is a failure mode this wiki had not seen.** Window 2026-10-21: **171 `completed` / 1 `failed`** (+ this audit `running`). The single failure is `dca8482e4f76` (**Daily Marketing Job Radar**, 07:00 weekday, `custom:omni-101` / `smart-route`), which died at 07:40:48 after **40 m 36 s** on `RuntimeError: Service temporarily unavailable: all targets were skipped by pre-dispatch filters reset after 19m 20s` — **not** the 503 resource-pressure string, which appears in no 2026-10-21 row → [[incidents/2026-10-21-omniroute-pre-dispatch-filter-skip]]. **Fleet unchanged: 28 jobs — 25 enabled / 3 disabled.****Non-clean `last_status` stays at 9**(7`error` / 2 `delivery_failed`) but the composition moved: **`08eb86f02d37` cleared** (its 08:30 slot ran clean; incident `08eb86_472c7b609c80`**resolved 2026-10-21 08:37:21**) while **`dca8482e4f76` re-entered** (streak 1). Survivors, with next run: `dca8482e4f76` (1, **10-22 07:00**), `0e519874d280` (2, **10-22 08:00**), `06a0a1c8704f` (1, **10-22 10:00**), `c8d52904e1d6` (1, **10-22 18:00**),`1b23c042e17d` (2, Fri 10-23 08:00), `c0a9fa314542` (1, Mon 10-26 09:00), `fad5a86979d4` (`delivery_failed`, Mon 10-26 03:00),`0e753e9720fd` (`error`, 0, 11-01 09:00),`066bb5191280` (`delivery_failed`, 11-01 02:00). **Incident ledger: 1 opened / 1 closed, unclosed stays 9** (8 `alerted` / 1 `detected`). **Pins:** 17 of the 28 jobs pin`custom:omni-101`(10`smart-route`, 6 `free-stack`, 1 `deepseek-paid-smart`), **7 carry no pin at all** (`provider`/`model` null → global default), 1 pins `custom:omni` / `smart-route`. **2026-10-20 settled at 169 `completed` / 1 `unknown`** — the gateway-orphaned run; the 10-21 audit's "168 / 0" was a mid-flight snapshot. **12 sessions in the window, all `cron` — no interactive sessions.** ⚠ **Infrastructure changed in-window: Hermes Agent was updated to `0.22.3` at 22:47 → 22:57 with its pre-update snapshot `skipped`** → [[systems/hermes-agent]], [[infrastructure/backups]].
- **Current as of 2026-10-23 00:1x AWST (`jobs.json.updated_at 2026-10-23T00:07:25+08:00`) — a zero-failure day, five incidents cleared, and a maintenance job that ran six times long and committed nothing.** Window 2026-10-22: **172 `completed` / 0 `failed`** (+ this audit `running`) — zero failures for the second time in threedays, with 4 runs already completed by 00:0x on 10-23. **A DB recount by `started_at` and by `claimed_at` puts 2026-10-21 at 168 / 1**, three below the 171 / 1 the 10-22 audit recorded; retention (floor 2026-10-08) does not explain the delta, which is logged here rather than papered over. **Fleet unchanged: 28 jobs — 25 enabled / 3 disabled.****Five incidents closed in-window and none opened**—`0e5198_be2291809e56` + `0e5198_ec2fac98a91d` (`0e519874d280`, closed 08:01:40), `06a0a1_9e0ff4eedc49` (10:14:34), `c8d529_e8c70250f9be` (18:17:59), `dca848_014affeb6954` (07:59:22) — so **unclosed rows fall 9 → 4** (3 `alerted` / 1 `detected`: `1b23c0_c35703f637b8`, `1b23c0_103cbe04a79b`, `c0a9fa_764558b5ae83`, `0e753e_e7f8194a7d06`). **Non-clean `last_status` falls 9 → 5:**`1b23c042e17d` (`error`, streak 2 — **next slot Fri 2026-10-23 08:00**), `c0a9fa314542` (`error`, 1 — Mon 10-26), `0e753e9720fd` (`error`, 0 — 11-01), `fad5a86979d4` (`delivery_failed` — Mon 10-26 03:00), `066bb5191280` (`delivery_failed` — 11-01). All four cleared rows had been red since **2026-10-15** (the `omni-101` 503 burst) and each ran clean on its first slot of 10-22. **Pins:** 11`custom:omni-101`/`smart-route`, 8 unpinned (`provider`/`model` null → global default), 6 `custom:omni-101` / `free-stack`, 2 `custom:omni` / `smart-route`, 1 `custom:omni-101` / `deepseek-paid-smart` (this audit). ⚠ **Stocktake anomaly:**`1a8e90ea2271` ran **03:00:28 → 04:38:19 (97 m 58 s)** against 16 m 01 s on 10-21, finished `completed` with delivery `suppressed` — and **committed nothing**. It left its regenerated `systems/skills-index.md` uncommitted (mtime 10-22 03:00, with a `## .Archive (41)` section reading **347**) and its diagnosed fixes unapplied → [[incidents/2026-08-15-skills-index-drift]]. **27 sessions in the window:** 15 `cron`, 3 `subagent` (the AI-Newsletter research fan-out), 6 `desktop` (voice-chat-backend research 11:28–11:32; phone-call tests 13:04 / 13:07 / 13:15, models `mimo-v2.5`, `or-inclusionai_ling-2_6-1t`, `minimax/minimax-m3.1-flash`), 2 `acp` (0 messages), 1 `oneshot` memory-capture probe. → [[systems/skills-index]], [[incidents/2026-08-15-skills-index-drift]]
- **Current as of 2026-10-22 00:1x AWST *(superseded by the 2026-10-23 entry above)* (`jobs.json.updated_at 2026-10-22T00:11:31+08:00`) — the fleet is clean except for one failure, and it is a failure mode this wiki had not seen.** Window 2026-10-21: **171 `completed` / 1 `failed`** (+ this audit `running`). The single failure is `dca8482e4f76` (**Daily Marketing Job Radar**, 07:00 weekday, `custom:omni-101` / `smart-route`), which died at 07:40:48 after **40 m 36 s** on `RuntimeError: Service temporarily unavailable: all targets were skipped by pre-dispatch filters reset after 19m 20s` — **not** the 503 resource-pressure string, which appears in no 2026-10-21 row → [[incidents/2026-10-21-omniroute-pre-dispatch-filter-skip]]. **Fleet unchanged: 28 jobs — 25 enabled / 3 disabled.****Non-clean `last_status` stays at 9** (7 `error` / 2 `delivery_failed`) but the composition moved: **`08eb86f02d37` cleared** (its 08:30 slot ran clean; incident `08eb86_472c7b609c80`**resolved 2026-10-21 08:37:21**) while **`dca8482e4f76` re-entered** (streak 1). Survivors, with next run: `dca8482e4f76` (1, **10-22 07:00**), `0e519874d280` (2, **10-22 08:00**), `06a0a1c8704f` (1, **10-22 10:00**), `c8d52904e1d6` (1, **10-22 18:00**), `1b23c042e17d` (2, Fri 10-23 08:00), `c0a9fa314542` (1, Mon 10-26 09:00), `fad5a86979d4` (`delivery_failed`, Mon 10-26 03:00), `0e753e9720fd` (`error`, 0, 11-01 09:00), `066bb5191280` (`delivery_failed`, 11-01 02:00). **Incident ledger: 1 opened / 1 closed, unclosed stays 9** (8 `alerted` / 1 `detected`). **Pins:** 17 of the 28 jobs pin `custom:omni-101` (10 `smart-route`, 6 `free-stack`, 1 `deepseek-paid-smart`), **7 carry no pin at all** (`provider`/`model` null → global default), 1 pins `custom:omni` / `smart-route`. **2026-10-20 settled at 169 `completed` / 1 `unknown`** — the gateway-orphaned run; the 10-21 audit's "168 / 0" was a mid-flight snapshot. **12 sessions in the window, all `cron` — no interactive sessions.** ⚠ **Infrastructure changed in-window: Hermes Agent was updated to `0.22.3` at 22:47 → 22:57 with its pre-update snapshot `skipped`** → [[systems/hermes-agent]], [[infrastructure/backups]].
- **⚠ Current as of 2026-10-21 00:0x AWST *(superseded by the 2026-10-22 entry above)* (`jobs.json.updated_at 2026-10-21T00:07:08+08:00`) — a clean day, one orphaned run, and one long-running silent failure.** Window 2026-10-20: **168 `completed` / 0 `failed`** (plus 1 `unknown` and this audit) — **the first zero-failure day since 2026-10-12**, ending a seven-day run of failures, and **the `omni-101` 503 fault did not recur** (zero rows carry the resource-pressure string; the last such failure remains 2026-10-19). **Zero incident rows opened and 4 closed** (`dca848_867d9caa96a8`, `85edf6_256c5e150bc4`, `ecc1fc_c9976ea24553`, `ecc1fc_21a72aa11cc4`), so **unclosed rows fall 13 → 9** (8 `alerted` / 1 `detected`). ⚠ **The exception is a failure this ledger does not record:**`08eb86f02d37` (Gmail Action Digest, 08:30) was claimed 08:31:03 and closed **`unknown`** when the gateway exited uncleanly — *“Scheduler restarted after this execution's owner exited before a durable terminal state”*; no delivery row exists for the slot and no incident row was raised → [[incidents/2026-10-20-gateway-unclean-exit-orphaned-run]]. **Non-clean `last_status` 12 → 9** (7 `error` / 2 `delivery_failed`); every `failure_streak` ≤ 2. **Fleet unchanged: 28 jobs — 25 enabled / 3 disabled.** ⚠ **Discovery — green row, dead job:**`e27da6c34bbb` (Config Backup, 11:00) reads `ok` while its Gitea push has failed **every night since 2026-08-08** (75 unpushed commits) → [[incidents/2026-08-08-gitea-config-backup-push-dead]]. **Model pins unchanged** — 20 of 28 jobs still pin OmniRoute — but the **8 unpinned jobs moved upstream to `nous` direct** with the global default change, so the fleet now has two upstream paths. **2 interactive sessions in the window** (`talaria-ios` tax work; a `desktop` turn that received the canned non-answer). → [[incidents/2026-10-16-omniroute-503-resource-pressure]]
- **⚠ Current as of 2026-10-21 00:0x AWST *(superseded by the 2026-10-22 entry above)* (`jobs.json.updated_at 2026-10-21T00:07:08+08:00`) — a clean day, one orphaned run, and one long-running silent failure.** Window 2026-10-20: **168 `completed` / 0 `failed`** (plus 1 `unknown` and this audit) — **the first zero-failure day since 2026-10-12**, ending a seven-day run of failures, and **the `omni-101` 503 fault did not recur** (zero rows carry the resource-pressure string; the last such failure remains 2026-10-19). **Zero incident rows opened and 4 closed** (`dca848_867d9caa96a8`, `85edf6_256c5e150bc4`, `ecc1fc_c9976ea24553`, `ecc1fc_21a72aa11cc4`), so **unclosed rows fall 13 → 9** (8 `alerted` / 1 `detected`). ⚠ **The exception is a failure this ledger does not record:**`08eb86f02d37` (Gmail Action Digest, 08:30) was claimed 08:31:03 and closed **`unknown`** when the gateway exited uncleanly — *“Scheduler restarted after this execution's owner exited before a durable terminal state”*; no delivery row exists for the slot and no incident row was raised → [[incidents/2026-10-20-gateway-unclean-exit-orphaned-run]]. **Non-clean `last_status` 12 → 9** (7 `error` / 2 `delivery_failed`); every `failure_streak` ≤ 2. **Fleet unchanged: 28 jobs — 25 enabled / 3 disabled.** ⚠ **Discovery — green row, dead job:**`e27da6c34bbb` (Config Backup, 11:00) reads `ok` while its Gitea push has failed **every night since 2026-08-08** (75 unpushed commits) → [[incidents/2026-08-08-gitea-config-backup-push-dead]]. **Model pins unchanged** — 20 of 28 jobs still pin OmniRoute — but the **8 unpinned jobs moved upstream to `nous` direct** with the global default change, so the fleet now has two upstream paths. **2 interactive sessions in the window** (`talaria-ios` tax work; a `desktop` turn that received the canned non-answer). → [[incidents/2026-10-16-omniroute-503-resource-pressure]]
- **Current as of 2026-10-20 00:09 AWST *(superseded by the 2026-10-21 entry above)* — the `omni-101` 503 fault is BACK, and it is now the fleet's defining risk.** Window 2026-10-19: **159 `completed` / 13 `failed`**, against 10-18's 168 / 1 and 10-17's 163 / 5. **9 genuine model-call failures across 8 jobs** all carry the identical `RuntimeError: HTTP 503: Service temporarily unavailable due to resource pressure. Retry shortly.` on `custom:omni-101`, in two clusters — **06:01** (one job) and **07:01 → 12:01** (seven more runs across seven jobs). The other **4 rows are derivative Mnemosyne Watchdog exits** (`96eb95ce12f2` at 06:16, 06:31, 12:16, 12:31) that merely quote the coverage monitor's error. This is the **second burst in four days** on the same node with the same blast radius, while `/api/health` read `200 {"status":"ok"}` throughout — so [[incidents/2026-10-16-omniroute-503-resource-pressure]] is back to **`status: active`** (it had been downgraded to `monitoring` 44 h earlier). **Fleet unchanged at 28 jobs — 25 enabled / 3 disabled.****Non-clean `last_status` 7 → 12** (10 `error` + 2 `delivery_failed`); **unclosed incident rows 13** (12 `alerted` / 1 `detected`) — 4 opened and 2 closed on 10-19. Every `failure_streak` still reads ≤ 2 and the 10-20 00:00 slots ran clean, so nothing is being suppressed yet. **15 sessions in the window, all `cron` — no interactive sessions.** → [[incidents/2026-10-16-omniroute-503-resource-pressure]]
- **Current as of 2026-10-20 00:09 AWST *(superseded by the 2026-10-21 entry above)* — the `omni-101` 503 fault is BACK, and it is now the fleet's defining risk.** Window 2026-10-19: **159 `completed` / 13 `failed`**, against 10-18's 168 / 1 and 10-17's 163 / 5. **9 genuine model-call failures across 8 jobs** all carry the identical `RuntimeError: HTTP 503: Service temporarily unavailable due to resource pressure. Retry shortly.` on `custom:omni-101`, in two clusters — **06:01** (one job) and **07:01 → 12:01** (seven more runs across seven jobs). The other **4 rows are derivative Mnemosyne Watchdog exits** (`96eb95ce12f2` at 06:16, 06:31, 12:16, 12:31) that merely quote the coverage monitor's error. This is the **second burst in four days** on the same node with the same blast radius, while `/api/health` read `200 {"status":"ok"}` throughout — so [[incidents/2026-10-16-omniroute-503-resource-pressure]] is back to **`status: active`** (it had been downgraded to `monitoring` 44 h earlier). **Fleet unchanged at 28 jobs — 25 enabled / 3 disabled.****Non-clean `last_status` 7 → 12** (10 `error` + 2 `delivery_failed`); **unclosed incident rows 13** (12 `alerted` / 1 `detected`) — 4 opened and 2 closed on 10-19. Every `failure_streak` still reads ≤ 2 and the 10-20 00:00 slots ran clean, so nothing is being suppressed yet. **15 sessions in the window, all `cron` — no interactive sessions.** → [[incidents/2026-10-16-omniroute-503-resource-pressure]]
- **Current as of 2026-10-19 00:20 AWST *(superseded by the 2026-10-20 entry above)* (`jobs.json.updated_at 2026-10-19T00:17:06+08:00`) — one failure in 24 h, and it was a failure mode this wiki had not recorded before.** Fleet **28 jobs — 25 enabled / 3 disabled** (unchanged). Window 2026-10-18: **168 `completed`, 1 `failed`** — the failure is `ecc1fc13655a` (FreshRSS Midday, 12:01:10 → 12:08:25), which **wrote and delivered a canned “Your request was not processed…” sentence to Telegram** (`delivery_outcome: delivered` at 12:08:11) before dying on `RuntimeError: Model output entered a repetition loop and the stream was cut mid-loop; discarding the degenerate response.` → [[incidents/2026-10-18-degenerate-stream-canned-substitute]]. **Non-clean `last_status`: 8 → 7** — `1a8e90ea2271` (10-18 03:01) and `cbfe434a3b5e` (this audit, 10-18 00:42) cleared on schedule, leaving `1b23c042e17d` (`error`, streak **2**, next **Fri 10-23 08:00**), `0e519874d280` (**Mon 10-19 08:00** — next to fire), `c8d52904e1d6` + `06a0a1c8704f` (Thu 10-22), `ecc1fc13655a` (**10-19 12:00**), and the monthlies `0e753e9720fd` + `066bb5191280` (both 11-01). **Open incidents 8 → 7** (6 `alerted` / 1 `detected`); **1 opened, 2 resolved** in-window. Tallies: 10-17 = 163 / 5, **10-18 = 168 / 1**, 10-19 so far 5 / 0 (+ this audit `running`); **11 sessions in the window, all `cron` — no interactive sessions.** → [[incidents/2026-10-18-degenerate-stream-canned-substitute]]
- **Current as of 2026-10-19 00:20 AWST *(superseded by the 2026-10-20 entry above)* (`jobs.json.updated_at 2026-10-19T00:17:06+08:00`) — one failure in 24 h, and it was a failure mode this wiki had not recorded before.** Fleet **28 jobs — 25 enabled / 3 disabled** (unchanged). Window 2026-10-18: **168 `completed`, 1 `failed`** — the failure is `ecc1fc13655a` (FreshRSS Midday, 12:01:10 → 12:08:25), which **wrote and delivered a canned “Your request was not processed…” sentence to Telegram** (`delivery_outcome: delivered` at 12:08:11) before dying on `RuntimeError: Model output entered a repetition loop and the stream was cut mid-loop; discarding the degenerate response.` → [[incidents/2026-10-18-degenerate-stream-canned-substitute]]. **Non-clean `last_status`: 8 → 7** — `1a8e90ea2271` (10-18 03:01) and `cbfe434a3b5e` (this audit, 10-18 00:42) cleared on schedule, leaving `1b23c042e17d` (`error`, streak **2**, next **Fri 10-23 08:00**), `0e519874d280` (**Mon 10-19 08:00** — next to fire), `c8d52904e1d6` + `06a0a1c8704f` (Thu 10-22), `ecc1fc13655a` (**10-19 12:00**), and the monthlies `0e753e9720fd` + `066bb5191280` (both 11-01). **Open incidents 8 → 7** (6 `alerted` / 1 `detected`); **1 opened, 2 resolved** in-window. Tallies: 10-17 = 163 / 5, **10-18 = 168 / 1**, 10-19 so far 5 / 0 (+ this audit `running`); **11 sessions in the window, all `cron` — no interactive sessions.** → [[incidents/2026-10-18-degenerate-stream-canned-substitute]]
**Total installed: 348 skills** in `~/.hermes/skills/`.
**Total installed: 306 skills** in `~/.hermes/skills/`.
- **41 archived skills** are parked under `~/.hermes/skills/.archive/` and are excluded from this count — the loader does not expose them.
- The loader's own catalogue (`hermes skills list`, e.g. *34 hub-installed, 52 builtin, N local — N enabled*) is the authoritative loadable figure; this page is a filesystem inventory.
This page is auto-regenerated from the live skill inventory. Do not hand-edit.
This page is auto-regenerated from the live skill inventory. Do not hand-edit.
@@ -20,432 +23,515 @@ This page is auto-regenerated from the live skill inventory. Do not hand-edit.
- Category = directory structure under `~/.hermes/skills/` (`category/skill/SKILL.md`).
- Category = directory structure under `~/.hermes/skills/` (`category/skill/SKILL.md`).
- Skills directly under `~/.hermes/skills/` group under `uncategorized`.
- Skills directly under `~/.hermes/skills/` group under `uncategorized`.
- `openhands` — Delegate coding to OpenHands CLI (model-agnostic, LiteLLM).
- `openhands` — # OpenHands CLI
## Hermes (3)
## Chromadb (1)
- `hermes-personality-soulmd` — Edit and verify Hermes Agent's core personality/identity file (SOUL.md). Use when the user wants to change 'who Hermes i
- `chromadb` — # ChromaDB — Vector Database Operations for the Homelab
- `hermes-telegram-miniapp` — Deploy and operate the standalone Hermes Telegram Mini App repo onto a Hermes Agent LXC/VM. Covers the full CT460 workfl
- `hermes-tts-configuration` — Set up Hermes TTS: providers, voice, persona, config edits.
## Devops (109)
## Chromadb Preflight (1)
- `abyss-ios-pairing` — Pair an iPhone with this Hermes instance via The Abyss iOS app — Tailscale Funnel/Serve + API key + QR. Detects host OS
- `agent-observability` — Observability and evaluation for Hermes agent tasks: event logging, memory attribution, repair-memory tracking, failure
- `agent-stack-weekly-review` — Weekly health review for the Hermes+Mnemosyne agent stack: read-only data collection from governance, observability, Mne
- `browser-automation-recovery` — >
- `browser-backend-verification` — Verify active browser backend and CDP connectivity.
- `browser-tab-lifecycle-and-recovery` — >-
- `camofox-browser-automation` — Operate the Camofox headless browser service for authenticated scraping, periodic data extraction, and browser automatio
- `camofox-pitfalls` — Runtime pitfalls and edible fixes for the Camofox browser REST API on Anthony's homelab. Complements camofox-browser-aut
- `camoufox-remote-websocket` — Expose Camoufox as a remote Playwright WebSocket server for cross-host browser automation. Use when another agent needs
- `chat4000-ops` — Operate, diagnose, and pair the chat4000 Hermes plugin — the native iPhone/Mac app for Hermes. Covers CLI commands, devi
- `cloudflare-ops` — Operate Cloudflare services — DNS, Tunnels, R2 Object Storage (S3-compatible), API authentication, and domain management
- `codex-via-cliproxy` — Run the OpenAI Codex CLI (codex) on a machine where chatgpt.com/backend-api is geo-blocked (HTTP 451) or the stored OAut
- `courier-tracking-monitor` — Monitor a package/shipment; alert on change or delivery.
- `crawlee-batch-scrape` — Lightweight batch URL scraping with Crawlee. Extracts markdown/text from multiple URLs in parallel. Low resource footpri
- `cron-job-maintenance` — Debug, fix, and harden Hermes cron jobs — particularly multi-step autonomous jobs where models skip or incompletely exec
- `cross-profile-session-handoff` — Continue work from a dead or stuck Hermes specialist profile in the default/supervisor profile. Covers session dump insp
- `cua-computer-use` — Set up and operate cua-driver for GUI automation on headless Linux servers. Covers virtual display setup (Xtigervnc/Xvfb
- `cua-driver-lifecycle` — Manage cua-driver processes on Hermes VMs — diagnose runaway instances, implement idle-timeout wrappers, and configure H
- `dashboard-server-ops` — Operate, extend, and troubleshoot self-hosted Python HTTP API servers serving SPAs. Covers http.server-based dashboards,
- `docker-management` — Manage Docker containers, images, volumes, and Compose.
- `external-auth-script-integration` — Fix broken script authentication by redirecting to local helper scripts, preserving module APIs during rewrites, and ver
- `external-skill-integration` — Bring external skills, standalone CLI tools, and MCP server binaries from GitHub repos into Hermes — for Hermes skills u
- `fleet-wide-failure-triage` — Triage when many components fail at once from one cause.
- `free-web-tools` — >-
- `freshrss-intelligence-filter` — Intelligence-filter upgrade for FreshRSS midday cron: clusters articles by topic, flags urgent items, adds 'why this mat
- `gcp-vertex-ai-provider` — Configure Google Cloud Vertex AI as an inference provider in the homelab – save service account credentials, test connec
- `gmail-imap-email-monitor` — Cron-friendly Gmail IMAP pattern for Hermes Agent: find new messages by UID range, parse headers robustly, detect urgenc
- `google-apis` — Google Gmail, Calendar, Drive, Docs, Sheets tools.
- `google-contacts-birthday-sync` — Sync Facebook birthdays to contacts services (Google Contacts, iCloud). Cross-references Facebook birthday data with Goo
- `hermes-agent-install` — Install, reinstall, upgrade, and diagnose Hermes Agent installations. Covers the official installer (scripts/install.sh)
- `hermes-agent-ops` — Diagnose and fix problems with Hermes Agent's OWN runtime — the web UI (port 8787), the memory-ui dashboard plugin ('Hin
- `hermes-api-vault` — Manage Hermes plugin/agent API credentials stored in a Notion vault — query the vault, map entries to ~/.hermes/.env var
- `hermes-config-diagnostics` — >-
- `hermes-cron-operations` — Operate, diagnose, and fix Hermes Agent scheduled cron jobs — list jobs, inspect configs, debug failures, manage scripts
- `hermes-cron-patterns` — Reliable cron job patterns for Hermes Agent. Covers prompt hardening, step enforcement, verification loops, subagent del
- `hermes-gateway-config-loading` — >-
- `hermes-gateway-ops` — Operate, diagnose, and troubleshoot the Hermes Agent messaging gateway — platform connections, config management, servic
- `hermes-instance-migration` — Audit one Hermes Agent instance's skills/state and port a curated subset (skills, scripts, data, cookies) to another ins
- `hermes-instance-update` — Update and verify Hermes Agent instances across the estate.
- `hermes-mattermost-integration` — Configure/troubleshoot Mattermost adapter for Hermes.
- `hermes-memory-systems` — Evaluate, configure, migrate, and troubleshoot Hermes Agent memory providers. Covers all 9 official providers (Hindsight
- `hermes-multi-agent` — Design, configure, and troubleshoot multi-agent Hermes setups using profiles, delegate_task, Kanban, and Telegram bot se
- `hermes-plugin-live-verification` — Verify Hermes plugin enablement, hook registration, live event emission, dual-write consistency, and fallback evidence p
- `hermes-plugin-ops` — Install, configure, optimize, and troubleshoot Hermes Agent plugins — covers plugin installation via CLI, provider/routi
- `hermes-profile-ops` — Audit, diagnose, and optimize Hermes agent profiles. Covers systematic inspection of config.yaml, profile.yaml, SOUL.md,
- `hermes-provider-configuration` — Configure Hermes Agent inference providers and custom OpenAI-compatible endpoints. Use when the user asks to add/switch
- `hermes-routing-and-domains` — Route requests correctly across Hermes CT460's multi-service architecture. Covers Tailscale Serve, Cloudflare Tunnel, do
- `hermes-self-diagnosis` — >-
- `hermes-soul-forge` — Operate the Soul Forge plugin — manage community souls, import profile SOUL.md files from repos, verify profile sync, an
- `hermes-web-tools` — Use the web search + content-extraction capability already installed on this Hermes instance. Covers the built-in web_ex
- `hermes-webui-ops` — Operate, restart, and verify the standalone Hermes WebUI on port 8787 (NOT the official control UI — that's on 9119). Co
- `hindsight-docs` — Complete Hindsight documentation for AI agents. Use this to learn about Hindsight architecture, APIs, configuration, and
- `hindsight-import-mnemosyne` — Import memories from a Mnemosyne SQLite DB into self-hosted Hindsight (:8888). 6,658 chunks (episodic+gists+working+fact
- `hindsight-ops` — Hindsight memory system — self-hosted LIVE on :8888 (systemd hindsight.service; OmniRoute free-IA LLM + Google/Gemini em
- `hindsight-selfhost` — Self-host Hindsight bare-metal (no Docker) + official Control Plane dashboard. hindsight-all-slim + embedded pg0 + hybri
- `homarr-dashboard-management` — Manage Homarr v1 dashboard on runtipi — multi-board setup, permissions, schema validation, PVE integration, and master b
- `homarr-dashboard-ops` — >-
- `homelab-docker-update` — Audit, update, and maintain Docker containers running inside homelab LXCs/VMs — across runtipi, bare Docker hosts, and a
- `homelab-instance-migration` — Audit and migrate a Hermes Agent instance from a source (old/bloated) box to a target (lean) box. Covers skill auditing,
- `homelab-service-management` — Deploy, update, and troubleshoot services running as Docker containers or bare-metal on Anthony's homelab Tailscale node
- `homelab-virtual-desktop` — Build, persist, expose, and debug a viewable virtual desktop (TigerVNC + websockify + Tailscale serve + systemd) on the
- `ignis-ops` — Deploy and operate Ignis browser-native Obsidian on CT460.
- `job-board-scraping-reference` — Platform-specific scraping behaviors for Australian job boards (SEEK, Indeed, LinkedIn) including access patterns, salar
- `kanban-orchestrator` — Decomposition playbook + anti-temptation rules for an orchestrator profile routing work through Kanban. The "don't do th
- `kanban-worker` — Pitfalls, examples, and edge cases for Hermes Kanban workers. The lifecycle itself is auto-injected into every worker's
- `linux-systemd-desktop` — |
- `litellm-openrouter-key-pooling` — Add additional OpenRouter API keys to a LiteLLM proxy setup as a pool, without replacing the existing key. Use when the
- `litellm-proxy-ops` — Manage a running LiteLLM proxy instance via its REST admin API — add/remove models, register new providers, update proxy
- `llm-model-diagnostics` — Diagnose LLM model quality issues: why a model feels 'dumb', responses get truncated, reasoning models return empty cont
- `mattermost-ops` — Operate, diagnose, and maintain a self-hosted Mattermost server (native or LXC). Covers plugin management, Agents plugin
- `mcp-agent-integration` — Install and wire local stdio MCP servers (e.g. SearXNG AI Kit) into the user's coding agents — Claude Code, Goose, OpenC
- `memory-governance` — Govern durable memory writes for Hermes + Mnemosyne: audit admissions, lifecycle metadata, duplicate/contradiction check
- `mnemosyne-operations` — Operate, diagnose, and maintain Hermes Agent's native Mnemosyne memory system. Covers health checks, dream cycle consoli
- `multi-agent-git-ops` — Git push, rebase, and merge workflows for repos where multiple agents (Hermes, Nanobot, coding agents) commit. Covers Gi
- `nanobot-omniroute-skills` — Give Nanobot (CT333) access to OmniRoute's skills + tools via MCP. Approach A = add OmniRoute's streamable-HTTP MCP serv
- `nanobot-ops` — Operate, diagnose, and recover the Nanobot gateway service on Proxmox CT333 (clawtest). IP: 192.168.178.66. Access via P
- `nextcloud-ops` — Operate and recover Anthony's self-hosted Nextcloud (CT270, NextCloudPi). Covers the 503/outage diagnosis (maintenance m
- `obsidian-ops` — Operate, diagnose, and troubleshoot Anthony's Obsidian vault. For basic note read/write/search and evidence-based resear
- `omniroute-combo-troubleshooting` — Diagnose and fix OmniRoute model combos that disappear from the /v1/models catalog (and thus from the Hermes /model pick
- `omniroute-context-tuning` — >
- `omniroute-mcp-portal-cloudflare-auth` — Reference for authenticating Hermes to mcp-portal.martinwa.org/mcp via Cloudflare Access headers, plus Nando's config lo
- `omniroute-ops` — Operate, diagnose, and configure OmniRoute — Anthony's AI routing proxy. Covers database access, alias/routing diagnosti
- `omniroute-skill-caller` — Call any skill installed on the OmniRoute MCP server by NAME (e.g. "mattermost", "nanobot", "hermes-agent-framework") wi
- `omniroute-skill-execution` — Execute skills on an OmniRoute MCP server (the composure_statistic_module / omniroute endpoint). Covers the streamable-H
- `openclaw-ops` — Operate, diagnose, and configure the Open-Claw VM — Anthony's autonomous AI agent gateway node (QEMU VM 403). Covers con
- `pve-ops` — Proxmox VE operations via REST API and SSH. Covers guest management, disk reports, diagnostics, and credential-based API
- `pve-vm-console-recovery` — Recover a stuck Proxmox VM by injecting console keystrokes.
- `radar-jobs-dashboard` — Use when operating or extending the RADAR jobs dashboard.
- `readlater-ops` — Operate, deploy, and fix Anthony's self-hosted ReadLater app (Tony0410/readlater) — Next.js + SQLite in Docker on debian
- `runtipi-container-management` — Update, clean up, and troubleshoot Docker containers inside a Runtipi LXC running on a Proxmox VE host. Covers inventory
- `sdlc-review` — Review Kanban handoffs and route verified outcomes.
- `self-hosted-app-deploy` — Deploy, update, and troubleshoot self-hosted Docker applications on Tailscale-connected homelab VMs. Covers the full lif
- `solutions-first-agent-conduct` — How this agent must behave when a task hits a wall or a tool fails — lead with the attempt and the fix, not the limitati
- `tailscale-ops` — Operate and manage Tailscale configuration — Serve routes, Funnel, DNS, ACLs, and debugging. Covers adding/removing serv
- `tailscale-serve-route-management` — Manage Tailscale Serve routes on CT460 — the `--set-path` replaces-all-rules pitfall, the full route inventory, and the
- `tailscale-serve-routes` — Manage Tailscale Serve routes on CT460 atomically. Use when routes need adding, fixing, or verifying.
- `task-approach` — How the agent should approach any task given by Anthony — pre-work, skill loading, quality gates, and delivery standards
- `telegram-create-forum-topic` — Create a Telegram forum topic (thread) in a group/supergroup that has topics enabled. Wraps the Telegram Bot API's creat
- `tmux-config` — Configure and troubleshoot tmux — fix scroll/scrollback (mouse mode, history-limit, wheel bindings) and live-apply confi
- `tool-ecosystem-evaluation` — Evaluate external tools, plugins, skills, and integrations for Hermes Agent and Nanobot. Use when auditing free add-ons,
- `uber-eats-monitor` — Monitor an Uber Eats order and announce via Sonos when delivery is imminent. Supports timer-based tracking (confidence)
- `vision-model-validation` — Validate and select free vision-capable models via OmniRoute for Hermes auxiliary.vision tool.
- `watchers` — Poll RSS, JSON APIs, and GitHub with watermark dedup.
- `web-content-extraction` — >-
- `weekly-comms-tracker` — Track all weekly communications sent to Anthony to prevent repetition across weeks. SQLite-backed with content hashing f
- `wiki-maintenance` — Maintain the systems wiki for staleness and drift.
- `zeroclaw-ops` — Operate, diagnose, and configure ZeroClaw — Anthony's autonomous AI daemon on Proxmox CT333. Covers config editing (TOML
## Productivity (26)
## Chromadb Skills Rag (1)
- `airtable` — Airtable REST API via curl. Records CRUD, filters, upserts.
- `petdex` — Install and select animated petdex mascots for Hermes.
- `petdex-custom` — Create custom Hermes petdex pets by composing AI-generated or imported character art into spec-correct spritesheets. Com
- `powerpoint` — Create, read, edit .pptx decks with python-pptx.
- `product-price-monitor` — Watch product, flight, or listing prices; alert on target.
- `session-librarian` — Organize sessions by prompt: find, rename, archive, prune.
- `stratos-command-centre` — Run, verify, extend, and diagnose the Stratos command-centre dashboard and its backend.
- `teams-meeting-pipeline` — Teams meeting summaries, job replay, Graph subscriptions.
- `telegram-cron-table-formatting` — How to deliver tables/structured data to Telegram from a Hermes cron job (deliver='telegram') or regular chat message so
- `tui-widgets` — Author live widget apps for the Hermes TUI dock.
- `wikilink-audit` — Scan and fix broken [[wikilinks]] in an Obsidian vault or markdown knowledge base. Finds orphans, broken links, and temp
## Email (17)
- `agent-email-patterns` — Architecture patterns for AI agents that communicate over email -- why agents need dedicated inboxes rather than human e
- `agentmail` — Give the agent its own inbox: send and receive email.
- `agentmail-check-email` — Read, search, summarize, and triage AgentMail inboxes through the connected MCP server. Use for ANY request to look at,
- `agentmail-cli` — Operate AgentMail from a shell with the official CLI. Use when the user wants commands for listing or creating inboxes,
- `agentmail-manage-inboxes` — Create, list, inspect, update, or delete AgentMail inboxes through the connected MCP server. Use for ANY inbox lifecycle
- `agentmail-mcp` — Configure or troubleshoot the hosted AgentMail MCP server for Codex, Claude Code, Cursor, or another Streamable HTTP MCP
- `agentmail-rest-fallback` — AgentMail REST API operations for reading messages and downloading attachments when MCP tools are unavailable. Covers th
- `agentmail-sdk` — Deprecated alias of the agentmail skill, kept so existing installs and pinned URLs keep resolving. Prefer installing age
- `agentmail-send-email` — Draft, send, reply to, or forward email through the connected AgentMail MCP server. Use for ANY request to send, reply t
- `agentmail-toolkit` — Add AgentMail tools to agent frameworks with the TypeScript or Python AgentMail Toolkit. Use for Vercel AI SDK, LangChai
- `cron-email-briefing` — Produce structured email briefings from a cron job using IMAP (himalaya CLI). Covers the fallback workflow when no pre-i
- `email-for-ai-agents` — Deprecated alias of the agent-email-patterns skill, kept so existing installs and pinned URLs keep resolving. Prefer ins
- `apple-notes` — Manage Apple Notes via memo CLI: create, search, edit.
- `apple-reminders` — Apple Reminders via remindctl: add, list, complete.
- `findmy` — Track Apple devices/AirTags via FindMy.app on macOS.
- `imessage` — Send and receive iMessages/SMS via the imsg CLI on macOS.
## Frontend (4)
- `design-alignment` — >-
- `impeccable` — Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt,
- `landing-dashboard-reference` — >
- `pixel-perfect-iteration` — >-
## Media (4)
- `gif-search` — Search/download GIFs from Tenor via curl + jq.
- `heartmula` — HeartMuLa: Suno-like song generation from lyrics + tags.
- `songsee` — Audio spectrograms/features (mel, chroma, MFCC) via CLI.
- `youtube-content` — YouTube transcripts to summaries, threads, blogs.
## Note Taking (2)
- `obsidian-vault-from-git` — Clone, initialise, or integrate a pre-existing Obsidian vault stored on Gitea or GitHub. Covers credential retrieval, gi
- `obsidian-vault-manager` — Manage Obsidian vaults — vault discovery, git sync, .obsidian/ state files, plugin config, rename safety, and troublesho
## Web Development (2)
- `interactive-html-pages` — Build self-contained interactive HTML pages served from a local HTTP server. Covers single-file pages, fetching JSON fro
- `page-agent` — Embed an in-page natural-language GUI copilot in web apps.
## Smart Home (2)
- `openhue` — Control Philips Hue lights, scenes, rooms via OpenHue CLI.
- `sonos` — >
## Mcp (2)
- `fastmcp` — Build, test, and deploy Python MCP servers.
- `mcporter` — List, auth, and call MCP servers/tools from the terminal.
## Uncategorized (47)
- `chromadb` — >-
- `chromadb-preflight` — >
- `chromadb-skills-rag` — >
- `clean-architecture` — Structure software around the Dependency Rule: source code dependencies point inward from frameworks to use cases to ent
- `clean-code` — Write readable, maintainable code through disciplined naming, small functions, and clean error handling. Use when the us
- `cloudflare-tunnel-ops` — Operate Cloudflare Tunnels to expose internal services through public domains, bypassing network restrictions (e.g. work
- `cron-model-repair` — Repair cron jobs with dead model/provider pins.
- `ddia-systems` — Design data systems by understanding storage engines, replication, partitioning, transactions, and consistency models. U
- `devops`
- `devops-homelab-architecture-live-probe` — Build an accurate homelab architecture diagram by probing live infrastructure first, especially Proxmox VE and Tailscale
- `devops-proxmox-operations` — Use when operating Proxmox VE hosts, LXCs, VMs, storage, API tokens, noVNC consoles, DNS/hosts overrides, offline disk e
- `devops-service-management` — Umbrella for miscellaneous service operations skills — event logistics automation, Tailscale Serve, virtual desktop/X11,
- `devops-tailscale-web-routing` — Use when deploying or repairing self-hosted web apps behind Tailscale Serve/Funnel, especially path-prefix routing, stat
- `devops-telegram-bot-api-self-hosted` — |
- `domain-driven-design` — Model software around the business domain using bounded contexts, aggregates, and ubiquitous language. Use when the user
- `email-himalaya` — Himalaya CLI: IMAP/SMTP email from terminal. Covers configuration, cron automation, urgency assessment, and Gmail quirks
- `groktocrawl-deployment` — Self-hosted deployment guide for GroktoCrawl as a Firecrawl v2 replacement — resource requirements, port configuration,
- `hermes-api-vault` — Manage Hermes plugin/agent API credentials stored in a Notion vault — query the vault, map entries to ~/.hermes/.env var
- `hermes-desktop-plugins` — Write desktop app plugins that add UI panes and commands.
- `hermes-gateway-troubleshooting` — Gateway process management, MCP fleet debugging, post-migration verification, and config hardening for Hermes Agent LXC/
- `hermes-mcp-troubleshooting` — Diagnose and fix Hermes Agent MCP server connection issues, config corruption, startup banner false negatives, and relat
- `hermes-runtime-worker-diagnostics` — Use when many cron jobs fail at once, before job logic.
- `hermes-themes` — Author a Hermes color theme that skins every surface.
- `interactive-reference-pages` — Build clean, mobile-first, interactive HTML reference and checklist pages with proper dark-theme contrast, localStorage
- `ios-hig-design` — Design native iOS interfaces following Apple Human Interface Guidelines. Use when the user mentions "iPhone app", "iPad
- `job-hunting` — Comprehensive career transition coaching — research the user's background, discover opportunities, automate daily job-ra
- `job-hunting-assistance` — End-to-end job hunting support for a professional: research the user's background from available sources, locate existin
- `job-search-automation` — Operate, debug, and extend an autonomous job-board scraping pipeline (cron -> web_search_plus/web_extract_plus -> Camofo
- `job-tracker-enrichment` — Maintain and enrich Anthony's Notion "Job Opportunities 2026" Perth marketing/communications job tracker. Covers the bro
- `litprog`
- `markdown-to-anywhere` — On-demand cross-posting service: Anthony tells me to post a .md file to a platform, I format it appropriately and delive
- `officecli` — Create, analyze, proofread, and modify Office documents (.docx, .xlsx, .pptx) using the officecli CLI tool. Use when the
- `pragmatic-programmer` — Apply meta-principles of software craftsmanship: DRY, orthogonality, tracer bullets, and design by contract. Use when th
- `proxmox-operations` — Use when operating Proxmox VE hosts, LXCs, VMs, storage, API tokens, noVNC consoles, DNS/hosts overrides, offline disk e
- `refactoring-patterns` — Apply named refactoring transformations to improve code structure without changing behavior. Use when the user mentions
- `refactoring-ui` — Audit and fix visual hierarchy, spacing, color, and depth in web UIs. Use when the user mentions "my UI looks off" (or a
- `restaurant-menu-extraction` — Extract structured menu data (items, prices, ratings, delivery info) from food delivery sites like Uber Eats using Camof
- `software-design-philosophy` — Manage software complexity through deep modules, information hiding, and strategic programming. Use when the user mentio
- `system-design` — Design scalable distributed systems using structured approaches for load balancing, caching, database scaling, and messa
- `tailscale-serve-ops` — Operate Tailscale Serve to expose local web content through the tailnet. Covers route setup, backend management, path-pr
- `tech-ai-newsletter-digest` — Process and analyze tech/AI newsletter emails from Anthony's inbox using `himalaya` CLI, synthesizing key stories into s
- `ux-heuristics` — Evaluate and improve interface usability using heuristic analysis. Use when the user mentions "usability audit", "users
- `web-design-reviewer` — This skill enables visual inspection of websites running locally or remotely to identify and fix design issues. Triggers
- `web-typography` — Select, pair, and implement typefaces for web projects. Use when the user mentions "font pairing", "which typeface", "li
- `weekly-review` — Sunday 18:00 AWST cron that compiles a weekly review: job application stats from Notion, cron health from SQLite executi
- `working-with-legacy-code` — Safely change and test untested codebases using Feathers'' "Working Effectively with Legacy Code". Use when the user men
- `homelab-migration-planning` — Plan homelab service migrations across Proxmox hosts.
- `in-repo-documentation` — Use for README and in-repository wiki documentation.
- `jobs-dashboard-server-ops` — Operate, debug, and extend the Notion-sourced job tracker dashboard (port 9099). Covers API routes, cover letter serving
- `llm-proxy-streaming` — Use when an agent surfaces raw SSE chunks as errors.
- `media-download-ops` — Operate homelab torrent clients via Web API.
- `multi-agent-estate-knowledge` — Use for shared docs across multiple agent instances.
- `pre-flight-investigation` — Check the target and this box before installing anything.
- `pve-lxc-intrusion-detection` — Detect cryptominers on a Proxmox LXC via pct exec.
- `qwenpaw-agent-operations` — Troubleshoot QwenPaw agents and model routing.
- `rclone-mount-ops` — Fix rclone FUSE mount recursion and CPU issues on PVE.
- `security-incident-response` — Investigate and remediate homelab security incidents.
- `self-hosted-dashboard-deployment-verification` — Use when verifying dashboard deployments.
- `nanobot-ops` — > **Session detail / full diagnostic transcript:** see `references/nanobot-webui-auth-and-404-recovery.md` (covers the v0.2.x 404 root cause, the `tokenIssueSecret`-vs-`token` password trap, port m...
- `nextcloud-ops` — # Nextcloud Ops (CT270)
- `obsidian-ops` — \n# Obsidian Vault Operations\n\nAnthony's Obsidian vault at `/home/hermes/obsidian-vault`, git-synced to Gitea via `obsidian-git` plugin (auto-sync every 5 min).\n\n## Vault Sync Scripts\n\nTwo au...
# Group by the top-level category (first directory under skills_dir)
# Group by the top-level category (first directory under skills_dir)
categories={}
categories={}
@@ -102,6 +108,11 @@ lines = [
"",
"",
f"**Total installed: {sum(len(v) for v in categories.values())} skills** in `~/.hermes/skills/`.",
f"**Total installed: {sum(len(v) for v in categories.values())} skills** in `~/.hermes/skills/`.",
"",
"",
f"- **{len(archived_files)} archived skills** are parked under `~/.hermes/skills/.archive/` and are "
"excluded from this count — the loader does not expose them.",
"- The loader's own catalogue (`hermes skills list`, e.g. *34 hub-installed, 52 builtin, N local — "
"N enabled*) is the authoritative loadable figure; this page is a filesystem inventory.",
"",
"This page is auto-regenerated from the live skill inventory. Do not hand-edit.",
"This page is auto-regenerated from the live skill inventory. Do not hand-edit.",
"",
"",
"## Stock-take notes",
"## Stock-take notes",
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.