tq-forge-scan
Skill8 starsChanged 5 months agoMITView on GitHub
Scan every installed skill and detect coverage gaps — workflows you keep doing by hand that don't have a skill yet. Builds a coverage map from skill descriptions and recommends forge candidates. Use when asked for "/tq-forge-scan", "find skill gaps", "what's missing", or "scan for forge candidates".
In this repository17 other files it gives its agents.
- handoffSkill
- keep-warmSkill
- savingsSkill
- switchboardSkill
- tq-forge-agentSkill
- tq-forge-agentsSkill
- tq-forge-improve-allSkill
- tq-forge-improveSkill
- tq-forge-learnSkill
- tq-forge-listSkill
- tq-forge-promoteSkill
- tq-forge-queueSkill
- tq-forge-resumeSkill
- tq-forgeSkill
- tq-forge-skillSkill
- tq-forge-statusSkill
- tq-forge-testSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 46
Gather candidate gaps: Ask the user what repetitive workflows they've been doing by hand lately, or skim recent session history if …
- Distinctive: line 73
When unsure, ask before recommending a forge.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/tq-forge-scan — gap detection across the skill library
When to use
You suspect there's a workflow you've re-done by hand several times that should have been forged already. Or you want a coverage report before a forge spree.
Procedure
export TQ_FORGE_HOME="${TQ_FORGE_HOME:-$HOME/.tq-forge}"
source "${CLAUDE_PLUGIN_ROOT:-${TQ_FORGE_HOME:-$HOME/.tq-forge}/install}/scripts/common.sh" && tq_ensure_home- Enumerate every installed skill.
ls "$CLAUDE_SKILLS_DIR" 2>/dev/null find "$HOME/.claude/plugins" -maxdepth 4 -name SKILL.md 2>/dev/null | head -100 ls "$SANDBOX_SKILLS" "$SANDBOX_AGENTS" 2>/dev/null - Read each skill's
description:frontmatter to build a coverage map:
Group what's covered into categories (browser, design, security, deployment, research, ops, data, communication, etc.).for d in "$CLAUDE_SKILLS_DIR"/*/; do name=$(basename "$d") desc=$(awk '/^description:/{flag=1;next} /^[a-z-]+:/{flag=0} flag' "$d/SKILL.md" 2>/dev/null | head -3) echo "$name | $desc" done - Gather candidate gaps. Ask the user what repetitive workflows they've been doing by hand lately, or skim recent session history if available. Each recurring manual task with no matching skill is a candidate.
- For each candidate, classify:
- Already covered — a skill exists with a matching description. Drop.
- Partially covered — exists but the description is mismatched. Suggest
/tq-forge-improve <existing>. - Real gap — no skill matches. Suggest
/tq-forge "<intent>".
- Render a gap report:
🔍 tq-forge-scan — coverage report Installed: <N> skills, <M> agents Forgeable: <K> sandbox + <P> promoted ─────────────────────────────────────────── 🟢 Covered: browser-test, ship, design-audit, ... 🟡 Partial: notion-summary (description vague) 🔴 Gap candidates: 1. weekly-metrics-digest 2. invoice-reconcile
Pitfalls
- The scan is heuristic — it matches descriptions textually, so a skill with a
poorly-written
description:can look like a gap even when it covers the case. When unsure, ask before recommending a forge. - Plugin-provided skills (under
~/.claude/plugins) are usually not yours to edit — list them as covered/reference, not as improve candidates.
Verification
- The "Installed" count is in the same ballpark as
find "$CLAUDE_SKILLS_DIR" -name SKILL.md | wc -lplus plugin skills. - Each gap candidate is a workflow with no matching installed skill description.
Tags
tq-forge scan gap-detection read-only
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

