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 agentmods add commands/wei63w/pm-manager/pm-fixgit clone --depth 1 https://github.com/wei63w/pm-managerWhat 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 | $0.00022 | $0.00465 |
| Opus 5 | $0.00011 | $0.00233 |
| Sonnet 5 | $0.00004 | $0.00093 |
| Haiku 4.5 | $0.00002 | $0.00047 |
Grade A, and why
pm-fix 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 3d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- fix — 100% identical, 0 lines differ
What it actually says
User Input
$ARGUMENTS
Outline
先告诉用户（中文，置顶）：本次只分诊，不改代码。要把修复写进仓库，请你明确确认。
- Prefer current conversation paste as evidence (first-class; no need to save a file first); else
--path/ sources /.pm/inbox/stacks/. - Desensitize; write evidence JSON; optionally copy redacted snippet to inbox. Never persist secret originals.
- Classify severity (blocking/high/medium/low/suggestion). Vue / Vite 白屏、
Vue warn、Pinia、路由失败按同样规则分诊，不改代码。 - Append
bugs/findings.md+state/todo.mdentry with next action. - Charter compare for availability/success criteria with confidence only if charter.status=approved.
- Persist a desensitized dialogue note via
pm log --kind fix(intent + files the user named; elseunresolved— never invent paths). Thenpm journalif you need the timeline. - Summarize in Chinese: severity, suggested
/pm-next/ later/pm-done TODO-xxx. Repeat: 未确认不改业务代码。
Do not modify application code unless user confirms.
Shared Workflow (all /pm-* commands)
- Read
.pm/config/project.yaml（缺则建议/pm-init）。 - Map-first; incremental; conversation paste highest priority.
- Desensitize evidence. Never persist raw secrets.
- Draft PRD/charter MUST NOT be used as the baseline.
- Merge findings/todos; refresh overview (all blocking+high, no 3-item cap).
- Append audit.
- Closing:
_closing.md轻量档。
Design baseline: docs/prd.md + 宪章。
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.
- 3d ago First seen · 41 lines · 22 tokens per session scan A 99e5184f4ebc
pm-fix is a command published in the GitHub repository wei63w/pm-manager (96 stars, last pushed 3d ago), licensed MIT. It adds 22 tokens to every session and 465 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
webperf
Run a web performance audit via the web-performance-auditor persona.
spec
Start spec-driven development — write a structured specification before writing code.
install
Add skills from GitHub repos, git URLs, or local paths.
audit-rules
Browse, enable, disable, and customize audit rules.
init
First-time setup. Auto-detects installed AI CLIs and configures targets.
search
Discover and install skills from GitHub repositories.