Verify OpenSpec docs: fresh-context fact-checking of documentation
Fact-check finished OpenSpec documentation with a fresh-context subagent — re-run commands, validate claims against source, run openspec validate on examples, report findings with one-line fixes
- What
- Fact-check finished OpenSpec documentation with a fresh-context subagent — re-run commands, validate claims against source, run openspec validate on examples, report findings with one-line fixes
- Cost
- Free
- Needs
- an OpenSpec docs tree to verify and an agent that can spawn subagents — the skill is a verification workflow, not software to install
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @fission-ai's skill for verifying finished OpenSpec user documentation: a manually triggered verification pass (not part of the drafting loop — drafting belongs to write-openspec-docs, which is not bundled here). The method: scope the run to a page, one `##` section, or a list of changed claims; spawn one general-purpose subagent per unit with every placeholder filled and every path absolute. The reviewer plays two readers at once — a skeptical developer reading it cold and a fact-checker with the repo open — and checks, in order: Facts (re-run every terminal command shown — read-only commands anywhere, mutating ones only in a scratch directory; verify AI-surface commands like /opsx:propose against the skill sources; every output block must match actual output), Examples (any example spec must pass `openspec validate`), Structure (no re-explaining topics owned by other pages), Job fit (does the unit serve the page's one-line job?), Trust and slop (hype adjectives, claims without evidence, binary contrasts, colon reveals, em dashes, three punchy sentences in a row). Default is report, not rewrite: findings ranked most severe first, each with the quoted line and a one-line fix, plus exactly what was verified and how, and every claim that couldn't be verified and why. Fixes apply only on approval; two passes without converging means stop and escalate to the user. Honest caveats: designed for the OpenSpec repo's docs tree (the tree README's invariants and the repo's writing rules are part of what gets checked); the agent needs subagent capability. MIT licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Quick answer
Verify OpenSpec docs is a free Muse skill for verifying finished OpenSpec docs with a fresh-context subagent, re-running commands, validating claims, running openspec validate, reporting findings with one-line fixes. A manual verification pass, not part of drafting, scoped to a page, section, or changed claims, one subagent per unit, placeholders filled, paths absolute.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 18, 2026📌 Pinned version: commit 9111a76Step 0 — before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: an OpenSpec docs tree to verify and an agent that can spawn subagents — the skill is a verification workflow, not software to install Install "Verify OpenSpec docs: fresh-context fact-checking of documentation" for me. It teaches my agent @fission-ai's doc-verification workflow: scope the run to a page, section or claim list; spawn one fresh-context subagent per unit with absolute paths; the reviewer re-runs terminal commands (read-only commands anywhere, mutating ones only in scratch), runs openspec validate on example specs, and checks facts, examples, structure, job fit, and trust-and-slop — then reports findings ranked most severe first with quoted lines, one-line fixes, exactly what was verified and how, and every unverifiable claim. Report-only by default; apply fixes only on approval. MIT-licensed. Repository: https://raw.githubusercontent.com/fission-ai/openspec/9111a7654d7800391459431fff4eaf66e33a3d2e/.agents/skills/verify-openspec-docs/SKILL.md Pinned version: install the code at commit 9111a7654d7800391459431fff4eaf66e33a3d2e exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "verify-openspec-docs". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. run it only when you want a verification pass — drafting belongs to the write-openspec-docs skill, not bundled here; hand the agent the docs page or section to verify). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- 9111a76 · 2026-10-13
- Sandbox install test
- Install ran in our test sandbox · Nov 18, 2026 · Observed during the install test: registry.npmjs.org
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.