Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jerry7991/standup-ghost --skill doctorgit clone --depth 1 https://github.com/jerry7991/standup-ghostWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jerry7991/standup-ghost/doctor)<a href="https://agentmods.dev/skills/jerry7991/standup-ghost/doctor"><img src="https://agentmods.dev/badge/skills/jerry7991/standup-ghost/doctor.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00075 | $0.00743 |
| Opus 5 | $0.00037 | $0.00371 |
| Sonnet 5 | $0.00015 | $0.00149 |
| Haiku 4.5 | $0.00007 | $0.00074 |
Grade A, and why
doctor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 25 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Standup Ghost — doctor
Read-only except where it says so. Output: ONE table (component / status ✅⚠️❌ / remediation), then details only for non-green rows.
Probes
- Config parses; required keys per enabled card (
requires-config) present. Setup state — anything belowlive⇒ "setup incomplete at : <what's next>". - Sources & sinks — run each enabled card's Probe section. For flavor-variant cards, RE-RESOLVE (probe both candidate tool sets via ToolSearch); if the resolved flavor changed, update
config.flavorsand say🔧 healed. Unreviewed cards in play ⇒ flag distinctly.- Allowlist sanity (silent-failure class): run
node lib/allowlist.jsand confirm it lists the MCP tools of every enabled MCP-backed card. If it emits core tools only, the runner can't call Slack/Jira/Calendar and fails silently (file sink still receipts). Cause = cards dir not found; fix =STANDUP_GHOST_CARDS_DIR/defaultCardsDirresolving<runtime>/cards. - Connector liveness:
claude mcp list— any claude.ai connector at "Needs authentication" that an enabled card needs ⇒ ❌ with "open Claude Code →/mcpto re-auth "; it can't self-re-auth headless.
- Allowlist sanity (silent-failure class): run
- Scheduler liveness (the no-run failure class — nothing else can see it):
- Jobs loaded: attempt
launchctl print gui/$(id -u)/com.standup-ghost.daily(+.watchdog); on managed deny, hand the user the command and parse pasted output. - Plists present in
~/Library/LaunchAgents/and content matches a fresh render (schedule drift ⇒ offer regenerated paste). - Last-receipt age vs
behavior.alarm_threshold_days(askippedreceipt counts as alive — pause/absence days are honest liveness).
- Jobs loaded: attempt
- Runtime version drift:
~/.local/share/standup-ghost/runtime/VERSIONvs the installed plugin'splugin.json(viaplugin-rootpointer; refresh the pointer if the plugin moved). On drift ⇒ offer resync: BEFORE copying, show the old-vs-new generated allowlist diff; if it WIDENS, require explicit confirmation (a widened tool grant must never ride a routine update silently). Resync = recopy bundle + VERSION. - State health:
state-corruptmarker ⇒ show the quarantined file, offer rebuild (inspect quarantine, restore receipts it can prove, then clear the marker — delivery stays held until cleared). Pending backlog ⇒ per-day/per-sink table; ≥ threshold is loud. - Pause:
pausesets the marker (runs writeskippedreceipts — no alarms while away);resumeclears it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 25 lines · 75 tokens per session scan A 3de3f577012f
doctor is a skill published in the GitHub repository jerry7991/standup-ghost (2 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 743 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
unravel
Systematic debugging with a mandatory hypothesis preflight — no code changes before evidence; one change at a time, then a regression test. Use for any bug, test failure, or unexpected behavior.
autopsy
Blameless post-mortem / RCA — gathers facts, reconstructs a timeline, runs 5-Whys, defines action items, and writes a post-mortem for review. Use after an incident or outage.
accelerate
Systematic performance profiling and optimization with a mandatory baseline-first gate — measure before changing, re-measure after. Use when something is slow or you need to hit a latency/throughput target.
bug-triage
Investigate a Jira bug from its key to a root-cause verdict — classify bug-vs-feature, fan out read-only Explore agents across the repo, adversarially confirm the cause, and write a tasks/ -triage.md artifact a lean fix-planner can pick up. Orchestration only; read-only, never touches Jira, stops at verdict +…
mk:jira-analyst
Full Jira ticket context analysis via the jira-analyst agent — including media (attached images / PDFs / screenshots) and linked issues. Produces structured findings suitable for posting back as a Jira comment. Read-only. Triggers: 'analyze KEY', 'rca for KEY', 'describe KEY with media'. NOT for complexity scoring…
prove-pre-existing
Prove whether a build, compile, or test failure is caused by YOUR change or was already broken — by stashing your edits, re-running, and comparing. Read-only verdict; fixes nothing. Use when a build/test fails and the question is "is this my fault or pre-existing", or before blaming your diff. Triggers: is this my…