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 agents/f/prompts.chat/skill-managergit clone --depth 1 https://github.com/f/prompts.chatWhat 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 | $0.00026 | $0.00697 |
| Opus 5 | $0.00013 | $0.00349 |
| Sonnet 5 | $0.00005 | $0.00139 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
skill-manager 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 2d 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.
The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
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.
- 2d ago First seen · 102 lines · 26 tokens per session scan A 598d1635136e
skill-manager is an agent published in the GitHub repository f/prompts.chat (168,434 stars, last pushed yesterday), with no licence file. It adds 26 tokens to every session and 697 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 agents, from other repositories
markdown-lint-wizard
Use this agent when markdown files have been created, edited, or need formatting validation. This includes after writing or modifying any .md file, before commits that include markdown changes, or when the user explicitly requests markdown linting or formatting fixes.\n\nExamples:\n\n- Example 1:\n user: "Create a new…
azure-principal-architect
Expert Azure architecture guidance grounded in the Well-Architected Framework and current Microsoft Learn docs. Use to design new Azure workloads, weigh trade-offs across the five WAF pillars, select services and SKUs, plan multi-region or zero-trust topologies, or review an existing design against Microsoft best…
GPT 5 Beast Mode
Beast Mode 2.0: A powerful autonomous agent tuned specifically for GPT-5 that can solve complex problems by using tools, conducting research, and iterating until the problem is fully resolved.
playwright-capture
General-purpose Playwright screenshot agent. Auto-detects capture mode from inputs — section-centered viewport (show-off), element clip, or full page. Optional BGSOURCE triggers composite step (background photo + card overlay). Spawned by /show-off and any task needing screenshots.
code-reviewer
Generic code review agent. Reads CLAUDE.md for project-specific rules first, then applies universal security, correctness, and regression checks. Use immediately after writing or modifying code. Does not re-score architecture/maintainability — that is ck:quality's job; if a ck:quality report already exists for the…
plan-reviewer
Red-team reviewer for the /plan pipeline. Reads plan.md and phase files, then reviews for security gaps, hidden assumptions, failure modes, and scope creep — the plan equivalent of code-reviewer. Returns structured findings with ACCEPTED/NOTED verdicts.