Closes#5. Single authoring skill replaces the factory trio — one set of
standards, one script, one place for future governance rules. Routes to
create or improve flow based on context. Passes skill-audit with no findings.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016z2ZFYHQCex8yZAMVMTZzZ
- Rewrite assets/templates/tests/README.md with correct repo-root context and
SKILL_NAME placeholder in bats run command (was: `bats tests/`, wrong CWD)
- Add sed substitution for tests/README.md in new-skill.sh so SKILL_NAME is
replaced in scaffolded test docs; test added to new-skill.bats (red→green)
- Remove /write-eval reference from skill-improve SKILL.md; reword as direct
action since the skill does not exist in the kyberforge plugin
- Remove self-referential README.md rows from skill-audit and skill-improve
Files tables to match template and skill-write convention
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- validate.sh: move FAIL lines and failure summary to stdout; stderr
reserved for fatal script errors only (missing SKILL.md, bad args)
- skill-audit SKILL.md: replace concrete plugins/kyberforge/skills/...
example with abstract placeholder to fix meta-circularity
- skill-write SKILL.md: rephrase placeholder section-heading instruction
to remove embedded FILL IN: from a code span, clearing validator false positive
- skill-improve SKILL.md: wrap Step 2 root-cause example in a text fence
- skill-write assets/templates/README.md: update Files table to individual-
file rows so skill-audit can verify per-file coverage
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- skill-audit: remove contradictory skip clause in Step 2 (binary only)
- skill-improve: drop redundant "Fix the root" heading; keep specific directive
- skill-write: add section-rename guidance to body discipline step
- template SKILL.md: make Instructions heading an explicit FILL IN placeholder
- template README.md: remove "Invoke via your agent tool:" prefix to match convention
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- skill-audit: promote internal-working gotcha to ## Gotchas section;
narrow cross-plugin path check to exclude tests/ (dev-only, repo-level
deps are expected); require tests/README.md to declare that dependency
- skill-write/skill-improve: reorder descriptions to lead with "Use when..."
for consistency with skill-audit and the agentskills.io spec trigger pattern
- skill-write: add tests/README.md with self-contained bats setup instructions;
remove cross-skill reference to skill-audit's tests/README.md
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Both validate.bats and new-skill.bats constructed SCRIPT using
$(dirname "$BATS_TEST_FILENAME"), which resolves to tests/ — causing
every test to fail with file-not-found. Fixed to use
$BATS_TEST_DIRNAME/../scripts/ to reach the actual scripts/ directory.
Also clarifies that bats-support/bats-assert must be installed from the
repo root (not the skill root) to match where the tests load them from,
and adds text language tags to three output-template code blocks in
skill-audit's SKILL.md.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Move test infrastructure (validate.bats, new-skill.bats) from scripts/ to
tests/ — the spec defines scripts/ as executable code agents can run, so
test files don't belong there. Add tests/README.md placeholders with
bats-support dependency declaration.
Update skill-audit to permit tests/ and flag other unlisted directories,
add scripts/ purpose check, and add /skill-improve near-miss exclusion.
Update skill-improve and skill-write to cover tests/ in directory lists,
scaffold template, and authoring guidance.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace verbose three-pass output (punch list + priority table + fix
proposals) with a compact findings-only report: coverage line, findings
grouped by dimension with Why+Fix per entry, and a result block with
/skill-improve handoff. Suppress PASS lines — absence confirms pass.
Fix validate.bats executable bit.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Deletes the neuledge-context skill (.agents/skills/) and four kyberforge
plugin skills — marketplace-architect, plugin-create, promptfoo, and
write-agent — along with associated docs (adding-agents.md,
plugin-marketplace-architecture.md).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Applies evidence-based improvements to existing skills using signals
from grill sessions, audit reports, eval failures, and inline feedback.
Groups signals by root cause before editing to avoid per-symptom patching.
Hands off to /skill-audit on completion.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Two data errors in sources.md files:
- agentskillsio/sources.md: Contributing files used agentskills-* prefix
(e.g. agentskills-overview.md) but actual filenames have no prefix
(overview.md, specification.md, etc.)
- write-agent/references/sources.md: plugin-marketplace-architecture entry
used a machine-specific absolute path; replaced with repo-relative path
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Separates development-time reference material from live plugin docs.
agentskillsio/, agentsmd/, and examples/skill-write/ are not shipped with
the plugin — grouping them under research/ makes that boundary explicit.
README now lists all top-level files and explains what research/ is for.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Bats files moved up to scripts/ directly; tests/ subdirectory was non-spec
and created a directory structure not defined by agentskills.io.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Reorder skill-audit description to lead with 'Use when...' trigger (P3)
- Add concrete example to 'control calibration' body discipline check (P4)
- Add bats test files to README file tables for both skills
- Fix REPO_ROOT and SCRIPT paths in bats files after tests/ subdirectory removed
- Add three lessons: plugin cache isolation, spec-grounded rubrics, test file placement
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add references/description-quality.md and references/body-discipline.md to
skill-audit — condensed, rubric-focused extracts from the agentskills.io
specification docs. Both files are loaded conditionally via progressive
disclosure triggers added to Step 3 (Description and Body discipline
dimensions), so the agent consults the spec source when a finding is
borderline rather than relying solely on inline heuristics developed
during the skill-write authoring cycle.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Move validate.sh ownership to skill-audit/scripts/ — it is the canonical
structural validator; skill-write now delegates Step 5 to /skill-audit
- Add skill-write/references/scripts.md and deployment-modes.md for progressive
disclosure of package runner patterns and plugin cache isolation rules
- Fix skill-audit Step 1 cross-skill path reference (was repo-absolute, now
skill-relative); add manual fallback for sandboxed/Bash-denied contexts
- Scope Step 2 "read every file" to exclude binaries and unreferenced files
- Fix new-skill.sh next-steps output to reference /skill-audit instead of
the removed validate.sh
- Remove stale Dependencies section from skill-audit README; flip dependency
arrow — skill-write depends on skill-audit, not vice versa
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Moves the previous write-skill implementation to docs/examples/skill-write/write-skill/
for reference. The skill has been superseded by the spec-compliant skill-write rewrite.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Rewrote the skill authoring factory skill from scratch against the agentskills.io
specification. Renamed write-skill → skill-write (name now matches directory per spec).
skill-write:
- Full scaffold via new-skill.sh (annotated templates for SKILL.md, README.md,
scripts/, references/, assets/)
- validate.sh checks all spec constraints deterministically (name format/length,
description length, placeholder detection, line count, script rules)
- SKILL.md body includes description rules, body discipline, patterns, and scripts
guidance with "why" rationale throughout
- Templates usable standalone by agents and humans
skill-audit:
- Structural validation (via validate.sh) + seven qualitative dimensions
- Produces PASS/FAIL/SUGGESTION punch list with per-FAIL fix proposals
- Report-only: does not apply fixes
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds write-agent to plugins/kyberforge/skills/ — a factory skill parallel
to write-skill that authors Claude Code subagent definitions and cross-tool
plugin agents (Claude Code + GitHub Copilot CLI two-file pattern).
Includes research references (claude-code-agents.md, copilot-cli-agents.md,
cross-compat.md), three asset templates (subagent, plugin-agent-claude,
plugin-agent-copilot), eval coverage, and CATEGORIES.md updated to register
write-agent in the factory category per the conflict check finding.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Removes the project-scoped context MCP server entry with pinned library filters.
Context7 is now integrated directly into the research skill; the neuledge-context
project scope is no longer needed here.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds Context7 resolution before websearch for library/framework/API topics,
reducing reliance on web crawling for well-indexed libraries. Falls back to
websearch for unresolved libraries, concept topics, or when the user provides
starting URLs. Subagents are explicitly prohibited from calling Context7 to
prevent tool inheritance from producing duplicate or conflicting summaries.
Includes trigger and output evals for the Context7 path (resolves, fallback,
skipped for non-library topics), stale description and constraint fixes, and a
concrete "sufficient content" threshold.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Standalone /research skill that scans the codebase, discovers canonical
sources via websearch, reads and deepens in parallel via subagents, and
writes structured topic files + sources.md to an explicit output path.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- claude plugin validate --strict absent from standard test sweep
- gitleaks source/deployed config silent divergence risk
- shellcheck without -x blocks pre-commit on scripts using source
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gv5iNACZxumtF2k6TsK18q
gitleaks false positive (U1, Gitea issue #2):
- 'Token routing: Haiku/Sonnet/Opus' in ai-coding-factory-session.md:90
triggers generic-api-key on entropy match of "Token". Not a credential.
- ROADMAP.md now documents this pattern and triggers the same rule.
- Both .gitleaks.toml (deployed, read by hook) and scripts/gitleaks.toml
(source for setup-gitleaks.sh deploys) updated and aligned. Previously
out of sync — deployed file had docs/research/.* already; source did not.
ROADMAP.md: governance workstream Phase 2 expanded with 7 immediately-
actionable test suite gaps and 5 Chunk 6 CI gaps, all mapped to
CONTROLS.md requirements. Housekeeping updated with audit entry.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gv5iNACZxumtF2k6TsK18q
Per repo convention, .gitkeep files are removed when the directory is
first populated with real content. Both directories are awaiting their
first ARD and Bug Brief respectively (Chunk 3+ work).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gv5iNACZxumtF2k6TsK18q
claude plugin validate scans all .md files in agents/ as agent
definitions and warns on missing YAML frontmatter. The file was a
contributor guide, not an agent. Kyberforge ships no agents, so the
agents/ directory is now correctly empty.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gv5iNACZxumtF2k6TsK18q
Two related fixes exposed when install.sh was first staged post-audit:
1. shellcheck invocation in setup-hooks.sh lacked -x, causing SC1091
(info) to fire for any .sh file that sources another, blocking the
pre-commit hook on legitimate scripts.
2. The shellcheck source= directive in install.sh pointed to
'deploy-manifest.sh' (bare filename). With -x, shellcheck resolves
this from CWD (repo root), where the file doesn't exist. Updated to
'scripts/deploy-manifest.sh' — the correct repo-root-relative path.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gv5iNACZxumtF2k6TsK18q
- neuledge-context v1.2: document that --libs identifiers must include
version suffix verbatim from `context list` (e.g. name@latest);
add rebuild workflow for packages with bad crawl/low section count;
add failure cases for get_docs returning Package not found and
/reload-plugins not restarting stdio MCP processes
- .mcp.json: fix all --libs identifiers to include @latest suffix;
update claude-code-docs to @2.1.98 (rebuilt from GitHub, 636 sections
vs 9 from the previous bad web crawl)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
- Fix claude mcp add scope flag: -s user for global, -s project for
per-project --libs (--project flag does not exist)
- Project scope writes to .mcp.json (committed); not settings.local.json
- Document dual-scope pattern and expected conflict warning
- Document MCP tools not available mid-session after claude mcp add;
require /reload-plugins or new session
- Expand context add step with llms.txt-first workflow for registry gaps
(Anthropic, Claude Code, MCP docs not in public registry)
- Fix self-check: explicit scope flag required, context list for identifiers
- Add .mcp.json with project-scoped context serve --libs for this repo
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
Adds a complete cross-cutting skill to install, configure, and manage
@neuledge/context — a local-first MCP server that delivers version-specific
library docs to AI agents via SQLite FTS5.
Includes:
- SKILL.md with 9-step process: install, global MCP registration, per-project
--libs scoping, package management, auth, custom registry, upgrade, uninstall
- setup-neuledge-context.sh: pinned version install, idempotent version check
- secure-context-config.sh: chmod 600 on ~/.context/config.json after auth
- 13-case test suite covering both scripts (all pass)
- eval.yaml with 6 trigger tests and 4 output tests
- references/: context-cli-reference.md, http-mode.md, install-notes.md
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
trap '...' RETURN inside a function is NOT local to that function in bash
— it persists in the calling scope and fires on every subsequent function
return. After install_shellcheck set the trap, it fired again when
ensure_tool returned with $tmp_dir unbound, causing nounset abort.
Fix: change install functions from {} to () (subshell bodies) and use
trap EXIT instead of RETURN. The trap is now scoped to the subshell and
cannot leak to callers.
Also fixes double _os() call in install_jq and removes redundant local
declarations (subshells don't need them).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
setup-hooks.sh now installs shellcheck, jq, and yq if absent rather
than warning and continuing. Follows the same install pattern as
setup-gitleaks.sh (curl + install to TOOL_INSTALL_DIR=/usr/local/bin,
OS/arch detection, pinned versions).
Pinned versions: shellcheck 0.10.0, jq 1.7.1, yq 4.44.3.
The deployed pre-commit hook retains its runtime fallbacks as a safety
net for environments where tools are removed after setup.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
Two bugs in setup-hooks.sh:
1. awk+mv to replace a marker block created a 644 temp file, losing the
exec bit. chmod +x after every write_block call fixes this. Regression
test added to idempotency block.
2. set -euo pipefail in the deployed pre-commit hook caused grep to exit 1
when no files of a given type were staged, aborting the hook. Changed
all filter pipes to process substitution with || true so no-match is
handled gracefully.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
Adds setup-hooks.sh and check-manifests.sh as the deterministic
enforcement layer described in docs/research/governance_principles/CONTROLS.md.
- commit-msg: conventional commits pattern check (hard block)
- pre-commit: shellcheck on .sh, jq on .json, yq on .yaml/.yml,
SKILL.md frontmatter validation; optional tools degrade gracefully
- pre-push: full test suite + manifest cross-reference check
- check-manifests.sh: validates marketplace.json plugin sources,
plugin.json skill/hooks/mcpServers path references
- Marker-based blocks (idempotent, composable with gitleaks)
- 33 integration tests across two test scripts
Run scripts/setup-hooks.sh to install into any repo's .git/hooks/.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TP4EGbBg3XMcyF28Lx78XJ
No longer needed — kyberforge plugin-create provides a bundled template that
serves as the canonical scaffold reference. Remove hello-world from marketplace
manifests and update docs accordingly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Move all evals into each skill's own evals/ directory and test_scripts.sh into
marketplace-architect/scripts/ so test artefacts live alongside the code they test.
Also fix: hooks.json array→object, displayName title-case, marketplace.json owner
placeholders, and .github/plugin/marketplace.json metadata-wrapper schema divergence.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
