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 skills/benwu95/prospec/releasenpx skills add benwu95/prospec --skill releasegit clone --depth 1 https://github.com/benwu95/prospecWhat 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 | $0.00071 | $0.02818 |
| Opus 5 | $0.00036 | $0.01409 |
| Sonnet 5 | $0.00014 | $0.00564 |
| Haiku 4.5 | $0.00007 | $0.00282 |
Grade A, and why
release 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.
How it starts
The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Release Skill
Cut a new prospec release from main: bump the version strings, create the tag, and publish
GitHub release notes in the house format. Publishing the release (published event) triggers
.github/workflows/release.yml, which compiles the multi-platform standalone binaries and attaches
them as compressed assets — so this skill's job is the version bump + notes, not the build.
Language
The release notes, commit message, and everything this skill writes are in English — a release
is trust-zone / public-facing technical output, exempt from the Traditional Chinese change-artifact
rule (see prospec/CONSTITUTION.md → Language Policy, mirrored in AGENTS.md). Talk to the user in
their language; write the artifacts in English.
Preconditions — do not release until all hold
- All PRs for this version are merged to
mainandmainis checked out with a clean tree (git status). A release is cut frommain, never a feature branch. - Each change is archived (
/prospec-archive) — REQs graduated, Knowledge synced, summary written. The release notes' Dogfood section reports this; it must already be true. - Counts are already synced.
pnpm countsruns as the last step of each feature commit, not here — READMEs /index.md/ module READMEs already carry the right test/skill counts. Do not runpnpm countsduring the bump. (See memory: counts are generated bysync-counts.ts.) - Green locally:
pnpm test,pnpm typecheck,pnpm lint, andprospec check(14 drift checks). Capture the numbers — they feed the "Quality & tests" section. Coverage:coverage/on disk may be stale (predates the current version); re-runpnpm test:coveragefor a live figure. Readprospec checkWARNs by type, not as one bucket:knowledge-health(stale module README) — almost always a prior change's archive that changed a module's source but never touched its README (drift compares git commit times). Fix it before releasing: sync + commit that module README so the notes can honestly claim a green check. Then confirm the fix didn't tipknowledge-sizeover the L2 per-module budget (this project overrides it to 2000 tokens in.prospec.yaml; the shipped default is 1000) — if it did, trim the README back under budget (net-neutral) rather than accept a new WARN.knowledge-sizebudget WARN that pre-exists and is deliberate (the SC-002 signal) may be reported as such, mirroring how 0.5.2 framed it — do not thrash the README to chase it away.
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 · 196 lines · 71 tokens per session scan A 4aa8bd8ebb84
release is a skill published in the GitHub repository benwu95/prospec (5 stars, last pushed 2d ago), licensed MIT. It adds 71 tokens to every session and 2,818 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
ospec-goal
Create or advance a full OSpec goal using the current document, task graph, worker, review, and evidence workflow.
ospec
Document-driven OSpec workflow for initialization, change/goal routing, validation, archiving, and durable project knowledge.
ospec-change
Create or advance a lightweight OSpec change using the classic fast workflow.
review-spd
Findings-first code review workflow for AI coding agents. Use when the user asks to review uncommitted changes, commits in a date range, or a branch compared to the main branch / PR-style diff. Focuses on bugs, regressions, correctness risks, missing tests, security/data-safety issues, and other behavior-changing…
deep-discuss
结构化深度讨论 Skill，用于与用户进行多轮问题分析和方案设计。当用户描述一个问题现象、故障表现、 技术困惑、方案选择困难，或明确说"讨论一下"、"帮我分析"、"我遇到一个问题"、"你觉得怎么样"、 "帮我想想"、"我在纠结"时，必须使用本 skill。当用户提供了一段描述（可能附带截图）并期望深入分析 而非直接给答案时，也应触发本 skill。即使用户只是抛出一个现象描述没有明确提问，也要使用本 skill 来引导结构化思考。不要在简单的事实查询（"X是什么"）或明确的执行指令（"帮我写个脚本"）上触发。.
spec-driven-develop
Automates pre-development workflow for large-scale complex tasks. Use when the user mentions "rewrite", "migrate", "overhaul", "refactor entire project", "transform", "rebuild in [language]", "spec-driven", or describes any large-scale project transformation that requires planning before coding. Also triggers on…