socialadsmentor/sam-agentic-harness/session-selftest
Claude Code and agent skill by socialadsmentor/sam-agentic-harness
- community
Checking that the SKILL.md file is still in the repository.
Run a health check of the local harness at session start - memory index byte budgets, hook parse checks, plugin presence, task-pattern sanity
Install
npx vskill@latest install socialadsmentor/sam-agentic-harness/session-selftestSKILL.md
The same SKILL.md was published earlier as socialadsmentor/sam-agentic-harness/session-selftest.
Session Selftest
Mode A health check for a Claude Code harness install. Runs seven checks against one or more config trees and reports each OK / WARN / FAIL with one-line evidence. Exits 0 if no FAIL row, 1 otherwise, so it can gate CI/cron.
When to Activate
- At the start of a session, to confirm the harness (rules, skills, memory, scheduled
tasks, provider-tier pins) is in a healthy state before relying on it.
- After any install, upgrade, or multi-seat sync, to catch drift or a silently-disabled
plugin before it causes a downstream failure.
Usage
Config is entirely env-driven so the same script runs against any config tree, single- or multi-seat:
| Env var | Purpose | Default | |
|---|---|---|---|
SAM_HARNESS_TREES | Comma-separated config-tree directory names under $HOME, e.g. .claude,.claude-alt for a multi-seat setup. | .claude | |
SAM_HARNESS_MEMORY_CAP | Byte budget for each discovered MEMORY.md index file. | 25000 | |
SAM_HARNESS_PLUGIN_CHECK | A plugin@marketplace id to verify is enabled via claude plugin list. | unset - check skipped (OK) | |
SAM_HARNESS_TASK_PATTERN | Case-insensitive regex matched against scheduled-task names to decide which ones the scheduled-task check cares about. | unset - check skipped (OK) |
The seven checks
- Configured plugin enabled (
claude plugin list), ifSAM_HARNESS_PLUGIN_CHECKis set. - Memory-index byte budgets: each discovered
projects/*/memory/MEMORY.mdvs the cap. - Harness lint: retired-term grep across
rules/*.md+skills/*/SKILL.mdfiles. - Stray files in each tree's
rules/dir. - Scheduled-task health (
schtasks /query), ifSAM_HARNESS_TASK_PATTERNis set. - Config-tree sync drift:
rules/*.mdhash-compared across every configured tree. - Provider-tier context-window pins, if
scripts/checks/tier_env_ctx_window.cjsexists in
the primary tree (that check module ships separately; this simply wires it in when present).
Zero external dependencies. Windows-aware (spawns claude.exe / schtasks.exe directly, no shell, so paths with spaces are never mangled).
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/session-selftest/SKILL.md
Popularity
- Repository stars
- 4
- Repository forks
- 1
- 7-day trend score
- 8
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/socialadsmentor/sam-agentic-harness/session-selftest)