The previous round taught agent-audit's validator to permit disallowedTools but
left the skill that writes agents still forbidding it, in six places. Running
agent-author on any of the three fenced orchestrators would have stripped the
fence, and nothing would have caught it: the validator's allowlist is a permit
list, so an absent field passes. The template was the worst of them, since its
comment is copied verbatim into every new plugin-scope agent.
Where a list had to be restated it is now a pointer to field-inventory.md's
apm-agent-allowlist instead -- the same data validate.sh reads -- because a
roster copied into a template goes stale one step further out than the roster
itself. Where the text has to teach something it teaches the shape rule rather
than the exception: tools is an allowlist whose vocabulary differs per harness,
so verbatim copy makes one value wrong on one target; disallowedTools is a
denylist, where an unrecognised name denies nothing, so the worst case is a
missing fence rather than a wrongly granted capability.
ADR-0016's amendment claimed an unrecognised key is inert on Copilot while the
same ADR's Context says that behaviour is unconfirmed by research -- asserting
as settled the exact thing it flags as unknown, and justifying it with apm's
compile-time behaviour, which says nothing about Copilot's runtime. It is
rewritten into labelled tiers: confirmed for Claude Code with citations,
inferred by analogy for Copilot with the analogy's limits stated, unverified
where it is unverified, and the residual risk accepted explicitly with its
blast radius. It also no longer claims to restore a write sandbox: the denylist
does not deny Bash, which these agents inherit and legitimately need.
docs/hooks.md called the old root hooks.json a stale sync artifact -- it was
added in the plugin's creating commit and pointed at by main's Copilot manifest
-- and claimed both ecosystems now resolve hooks/hooks.json. Copilot does not:
its hooks field has no default and no compiled manifest declares one, so it
resolves nothing. Recorded as the gap it is, with re-injection noted as a
follow-up rather than asserted away. Its event list is marked partial.
Also: new-agent.bats asserted a hardcoded four-field allowlist and would have
rejected a scaffolded agent carrying the field the ADR now blesses; it reads
field-inventory.md too. And ADR-0016's premise that Claude's tools: is
space-separated was wrong -- it takes a comma-separated string or a YAML list.
The incompatibility with Copilot is the vocabulary, not the punctuation.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X7GvKuJfy2WrdBmUttV4DT
AGENTS.md told an offline agent to push with SKIP=apm-marketplace-check and
asserted that hook was "the only one whose failure mode is 'no network'".
Running all 12 pre-push hooks under a network namespace shows two fail, for
one shared cause: apm-pack-check-clean resolves the same remote entry. An
exact pin does not remove the ls-remote, so both hooks are named now.
AGENTS.md also said everything in a plugin root except .apm/ is generated.
Plugin roots carry hand-authored README.md, docs/, bin/, sources.md and
.mcp.json, so an agent would hunt for an .apm/ source that does not exist or
refuse the edit. The rule is positional: immunity belongs to the plugin root,
and anything inside a mirrored directory is still rm -rf'd.
ADR-0017 said apm strips a hooks field. The real loop is (agents, skills,
commands, instructions) -- hooks absent, instructions never mentioned -- and
it can never fire, because synthesize_plugin_json_from_apm_yml only emits the
eight identity fields. The decision stands; the mechanism was overstated. Its
mcpServers amendment is rewritten for the pointer payload and now records the
real reason: inlining bypassed apm's credential sanitizer.
ADR-0015's owner.email and version-pin passages are corrected against the apm
source, and ADR-0016 gains the disallowedTools amendment. agent-audit's
allowlist is data, so it gains disallowedTools too -- the ADR and the
validator that enforces it had come apart.
architecture.md described a root CLAUDE.md that imports two files (it imports
one, plus an RTK block) and pointed at an ADR index that does not exist.
Seven skill READMEs listed tests/ files the mirror strips, promising installed
users files their install lacks; those rows are marked source-only, with the
depth-4 template tests explicitly called out as surviving. And
plugins/kyberforge/hooks/README.md, deleted during the conversion and
preserved nowhere, is restored to a path the mirror does not own -- verified
by running a sync against a scratch copy.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X7GvKuJfy2WrdBmUttV4DT
Documentation asserted things the code no longer did, which is how several of this
round's defects survived three review passes.
- `docs/spec/architecture.md` still described the pre-APM model — "each plugin has a
`plugin.json` manifest" — when no such file exists. AGENTS.md routes agents here
for the current structure, so it was live drift, not archive. Also cited ADR-0012
where ADR-0003 is meant.
- AGENTS.md never said `.apm/` is the sole hand-edited source, while the generated
mirror is byte-identical with no marker — an agent reading only AGENTS.md would
edit the mirror and find out at pre-push. It also omitted that `apm` and `jq` are
now required to push, unlike the `vale` bullet that exists for exactly that reason,
and understated the pre-push gate. `apm marketplace check` hard-fails offline and
`--offline` does not help, so the `SKIP=` escape hatch is documented rather than
new machinery built.
- ADR-0017 carried hardcoded line numbers that drifted twice within one PR, and said
`mcpServers` reinjection was real-syncs-only where the script's own header says
both modes. Line numbers dropped rather than corrected.
- ADR-0015 asserted issue #90 was closed. It is open. Reworded to defer to the
issue's own state rather than depend on someone remembering to close it.
- ADR-0014 said both audit skills pass `--config`; both SKILL.md files say pass none,
and passing one fails with E100. The ADR was wrong.
- ADR-0007 cross-referenced an archived ADR-0011 whose number was later reused.
Disambiguated with a note — renumbering a published ADR rewrites an immutable
record and breaks every existing citation.
- CONTEXT.md claimed `lint-runner` is report-only via a missing `Edit` tool.
Plugin-scope APM agents cannot express `tools:` (ADR-0016), so it is report-only by
instruction now; ADR-0016 accepted that regression but CONTEXT.md was not updated.
- Recorded two undocumented consequences of the conversion: `displayName` dropped
from all six compiled manifests and `owner.email` from the marketplace, and
`mattpocock-skills` silently version-pinned and manually maintained.
- Wrote the version-bump policy issue #90 specified but never delivered. Its previous
carrier, `marketplace-author`, was deleted in this PR, leaving the per-package rule
documented nowhere.
- README fixes: kyberforge listed skills belonging to the `git` plugin and a deleted
hooks path, `bin` advertised an agents directory it never had, and the docs index
listed files that do not exist.
Refs: #90
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X7GvKuJfy2WrdBmUttV4DT
Validated the plugin-content-mirror fix (issue #90) against apm's own
packing/CI documentation and source: no apm-native mechanism replaces the
mirror script (apm's bundler treats .apm/ and root convention dirs as
mutually exclusive, by design), but the investigation surfaced a real,
separate gap -- this repo ran zero apm-native audit/check commands in CI,
relying entirely on custom scripts and Claude Code's own client-side
validator.
Add three pre-push hooks matching apm's documented producer CI pattern:
- apm marketplace check: validates every marketplace.packages[] entry
resolves, including live network reachability for remote refs -- a
blind spot check-manifests.sh explicitly skips (local sources only).
- apm audit --ci: apm's own lockfile/policy/hidden-content integrity gate.
- apm pack --check-versions --check-clean: closes issue #90's deferred
item 3 (a check-clean-equivalent gate) using apm's native flag instead
of bespoke drift logic, verifying .claude-plugin/marketplace.json still
matches what apm.yml + .apm/ would currently generate.
All three are network-tolerant and whole-repo in scope, so they belong at
pre-push alongside check-manifests/check-plugin-content-sync/
validate-plugins -- not pre-commit, which stays fast/offline/per-file.
Documented the packing/bundling/releasing/CI findings in
docs/research/docs/microsoft-apm/releasing.md (new) and extended
testing-and-validation.md with the apm-action wrapper and its documented
CI patterns, sourced from Context7 and cross-checked against the
installed apm-cli 0.28.0 package directly.
Refs: #90
The file's body cites APM's Python source directly (apm_cli/models/validation.py,
apm_cli/primitives/models.py) to support its "no dedicated Agent validation"
claim, but source_keys only listed context7-microsoft-apm — the apm-github-repo
source_key was missing, and sources.md's apm-github-repo entry didn't list this
file under Contributing files either. The three sibling schema files from the
same commit (prompt/instructions/hooks) all cited both sources correctly.
Found via post-implementation review of issue #89.
skill-author/agent-author's #89 retarget needs to know exactly how each
.apm/ primitive compiles to Claude Code and Copilot CLI output. The
existing microsoft-apm corpus only had a full schema for skills and one
minimal example for agents, and nothing for prompts/instructions/hooks.
Deepened via APM's own Python source (not just docs) where prose was
thin. Key finding for #89: agents have no per-target integrator, so
apm compile does a naive verbatim copy to both Claude and Copilot,
unlike prompts/instructions/hooks which each get real per-target
reconstruction. That means the agent primitive's `tools:` field can't
express both harnesses' incompatible vocabularies at once — a real
upstream gap, not something we can schema our way around.
Re-review (comment 24) of fix commit e16c3dc found six new issues, mostly
introduced by that fix commit itself: a dangling reference to a Hard Rule
bullet the same commit deleted (apm-orchestrate.md/.agent.md Process step
2 still named "secret indirection"), and an ADR-0015 Decision bullet that
claimed "this ADR does not update CONTEXT.md" while the same commit had
just added a forward-pointer sentence to CONTEXT.md's Plugin/Plugin
marketplace entries. Both reworded to match what actually happened.
apm-install's APM_INSTALL_DIR escape-hatch example dropped the curl pipe
entirely (`APM_INSTALL_DIR=... sh` with nothing piped into it) — fixed in
both apm-install/SKILL.md and the installation.md research doc, verified
against the upstream Microsoft APM docs via Context7.
Neither apm-workflow nor apm-orchestrate routed to plain `apm install
[PACKAGE_REF]`, the CLI command that actually resolves/fetches
dependencies declared in apm.yml — apm-install only bootstraps the apm
binary/runtime, not per-package deps. Added a 5th "install" dispatch
action to apm-workflow (new references/install.md, SKILL.md table row,
README usage/files sync, sources.md provenance entry) and a matching
"install" operation group on apm-orchestrate so it can route there.
configure.md's apm.yml schema block was also missing the "legacy singular
`target:` CSV form is still accepted" caveat its sibling research doc
documents for the same field — added for consistency.
The sixth finding (paired .md/.agent.md Output-contract disagreement) was
checked against git-orchestrate and gitea-orchestrate's existing pairs and
found to match established repo convention (JSON schema in .md, prose
summary without the enum in .agent.md) — left unchanged as a false
positive rather than "fixed."
kyberforge bumped 1.3.0 -> 1.3.1 via agent-author's normal improve flow.
Refs: #91
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0186ERbyACLRuRxPRnqwpa4m
Capture Microsoft's Agent Package Manager (APM) — overview, install,
config, CLI reference, registries/marketplace, monorepo shapes,
testing/validation, troubleshooting, and examples — as structured
reference docs under plugins/kyberforge/docs/research/docs/microsoft-apm/.
Lays the groundwork for issue #88 (build agents/skills to execute a
marketplace-to-APM conversion of this repo).
Audits a target repo's AGENTS.md file(s) for embedded secrets, structural
completeness against the agents.md common-sections checklist, and drift
(referenced commands/paths that no longer resolve). First active skill in
the core plugin — kyberforge is scoped to marketplace-factory meta-tooling,
not generic target-repo documentation (see ADR-0012). Moves the agentsmd
research corpus from plugins/kyberforge to plugins/core to match.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Fills a gap needed for the planned forge orchestrator skill: neither doc set
previously stated when to build a skill vs a subagent/custom agent. Sourced
via context7 against the same libraries already recorded in each sources.md.
Also confirms agentskills.io's spec is runtime-agnostic and defines no agent
concept, so it has no bearing on this decision by design.
## Why
The existing kyberforge research covered skills vs agents (ADR-0010),
Copilot track selection, and agent scope hierarchy — but had no documented
basis for three practical decisions: when to use MCP servers vs skills vs
agents, when hooks are the right tool vs skills, and when bin/ is
appropriate vs scripts/ inside a skill directory. Without this, the
plugin-author and agent-author skills have no research backing for those
choices.
## Implementation Notes
Sourced from official Claude Code plugin docs (code.claude.com), official
GitHub Copilot CLI docs (docs.github.com), and the Copilot
customization-cheat-sheet and comparing-cli-features pages — the latter
containing the only official "putting it together" decision table across
components. Context7 MCP was the primary retrieval mechanism; web reads
deepened the hooks and bin/ content.
Three files produced:
- overview.md — mental model, component roles, cross-provider portability
- decision-guide.md — explicit decision tables for all three gaps,
including anti-patterns and Copilot surface support matrix
- hooks.md — full lifecycle event reference, stdin/stdout schema,
hookSpecificOutput per event, plugin scope restriction
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0147vXtL5sP6vorDdqXGJJU9
## Why
The initial research pass focused on agent definitions. An audit identified
critical and minor gaps in both doc sets around plugin publishing, marketplace
registration, and the Copilot extensibility model.
## Implementation Notes
Claude Code gaps filled: end-to-end publish walkthrough (scaffold → validate →
tag → host → register), CLI vs in-session command surface equivalence, all six
marketplace source URL formats, plugin update/upgrade lifecycle, interactive
plugin manager UI, private marketplace auth, `commands` vs `skills/` distinction.
Copilot gaps filled: discovered that the GitHub App-based Copilot Extensions
track was sunset November 2025. Created copilot-extensions.md as historical
reference (deprecated, with MCP servers as the current replacement path).
Documented the agent vs. skillset extension type distinction, OAuth install
flow, and VS Code Chat Participants as the surviving @mention mechanism.
overview.md updated with a "Which Track to Use" decision table covering all
four active tracks plus the deprecated one.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
## Why
Parallel research track to the Claude Code plugins research already
committed. Needed to understand GitHub Copilot's extensibility model
before designing cross-tool plugin compatibility for the kyberforge
plugin system.
## Implementation Notes
Three distinct Copilot extension tracks are covered: CLI plugins
(plugin.json + marketplaces), cloud/IDE custom agents (frontmatter .md
files committed to repos), and the SDK programmatic API. The SDK track
got its own topic file (sdk.md) because the content doesn't fit neatly
into the default topic list. Sources include Context7 (/websites/github_en_copilot),
both user-provided reference URLs, and four additional deepened pages.
## Impact
Provides a reference baseline for evaluating .claude-plugin/ / plugin.json
compatibility between Claude Code and Copilot CLI — the two formats share
a manifest discovery path and the strict:false field enables cross-tool
plugin distribution.
---
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Research from official docs (code.claude.com) and Context7, covering
plugin manifest schema, agent definition frontmatter spec, scope
priority, marketplace distribution, and plugin subagent restrictions.
Foundation for writing an agent-author skill.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds the gitea skill to kyberforge — a dispatch skill for managing
Defame1297/holocron via Gitea MCP from within Claude Code. Covers
issues, PRs, milestones, labels, branches, and status. Owner/repo
derived from git remote at runtime; no config required.
Also adds research docs (api-reference, data-model, examples,
overview, troubleshooting) and token-access reference used during
authoring and available for runtime scope lookups.
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>
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>
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>
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
Moves create-plugin, marketplace-architect, write-skill, and write-eval
from canonical .agents/skills/ into plugins/kyberforge/skills/, along
with all bundled sub-files, evals, and the plugin-marketplace-architecture
research doc. Bundles templates/plugin/ into create-plugin/assets/plugin-template/
so the skill is self-contained after install-time caching. Removes
templates/plugin/ and docs/research/plugin-marketplace-architecture.md
from the repo root as they are now exclusively in the plugin.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Scaffolds the kyberforge marketplace management plugin (create-plugin and
marketplace-architect skills will live here). Renames the plugin-create
skill directory to create-plugin to match the canonical kebab-order convention.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
