The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Phases Agents listing page.
English · Français

A local MCP server that discovers, validates and selects skills deterministically. Python standard library only, no runtime dependencies.
One server. Five tools. Nothing executed behind your back.
AI agents improvise. Ask the same question twice and you get two different plans. That is fine for brainstorming, and unacceptable for audit and compliance work.
phases-agents removes the improvisation. It profiles a local project, validates a catalogue of skills against a strict contract, and returns a plan that can be replayed. Same target, same catalogue, same parameters, same decision.

The server selects and exposes. The calling model reads the selected skills and decides what to do with them, using its own tools. The server never executes a skill.
The modules live in src/phases_agents/.
| File | Role |
|---|---|
validator.py | official contracts and validated snapshots |
skill_loader.py | bounded local discovery |
skill_runtime.py | trusted roots and verified cache |
skill_types.py | immutable types and limits |
registry.py | validated, immutable registry |
detector.py | local profile of the target |
planner.py | deterministic selection and ordering |
server.py | JSON-RPC/MCP transport |
capabilities.py | client capability vocabulary |
profile_facts.py | versioned profile-fact vocabulary |
skill_gaps.py | gap rules (skills_missing) |
The normative contract lives in src/phases_agents/core/SKILLS_CONTRACT.md
(French).
An example package ships in examples/skills/. Three steps produce a real
plan.
Create skills-roots.json pointing at the example root:
pip install -e . installs the phases-agents command. Without installing,
the same server starts with PYTHONPATH=src python -m phases_agents.server.
The server reads JSON-RPC line by line on standard input. A
phases_agents_plan call against a Python project then selects
hello-python:
Two plan formats coexist. "B3" names the versioned format, not a server
version; its official schema is
src/phases_agents/core/PLAN_B3_SCHEMA.json. Use it for new
work. Without plan_version, the legacy format returns a flat list of steps;
it is kept so existing callers do not break, and will be deprecated before
removal. client_capabilities is only accepted in B3, since declaring what
the client can do only makes sense in that format.
Claude Code (.mcp.json at the root of your project):
Codex uses the same command/arguments pair in its own configuration file. No token and no environment variable is required.
today is injected rather than read from a clock, so every call is
replayable. get_skill takes an identifier, never a path, and its content
comes from the validated snapshot. Absolute paths and detected secrets are
masked in public output. Any encoded JSON-RPC response stays under 1 MiB.
The first call builds the validated registry. Warm calls verify metadata
without re-reading contents. refresh_skills forces a rebuild.
Each package is a direct child of a root and contains at least:
The fastest way to start is to copy examples/skills/hello-python/ and rename
the identifier.
Five keys are allowed. All optional, all checked when present.
| Key | Constraint |
|---|---|
name | must equal phases.json.id |
description | bounded free text |
version | must equal phases.json.version |
owner | free author identity, no invisible characters |
license | Apache-2.0, MIT, BSD-2-Clause or BSD-3-Clause |
Each is a Markdown heading (##), in any order. Section titles are French,
because they belong to the contract; the body is yours to write in any
language.
Loi centrale · Ce que ce skill fait · Ce que ce skill ne fait pas ·
Conditions d'activation · Conditions d'exclusion ·
Capacites necessaires · Interdictions · Methode d'audit ·
Contrat de preuve · Format de sortie · Conditions de blocage ·
Limites connues · Exemples d'entree · Exemple de sortie attendue
All required: schema_version, id, version, title, description,
domain, project_types, platforms, activation, exclusions,
requires_capabilities, optional_capabilities, forbidden_capabilities,
execution_mode, human_approval, output_schema, rules_path,
references_path, scripts_path, tests_path, files.
output_schema uses the symbolic form core:SCHEMA_NAME.json.
project_types must intersect what the detector can emit: apk, python,
skill_package, solana, web.
activation.any uses profile facts: collects_personal_data, has_api,
has_apk, has_authentication, has_database, has_ecommerce,
has_eu_context, has_file_upload, has_javascript, has_python,
has_rust, has_skill_packages, has_solana, has_source_code,
has_typescript, has_web, uses_ai, uses_payments.
requires_capabilities, optional_capabilities and
forbidden_capabilities use: browser, dependency_installation,
filesystem_read, filesystem_search, filesystem_write, human_question,
shell, target_code_execution, web.
Provided capabilities are an open vocabulary: each catalogue names what
it brings, and only the shape is enforced (^[a-z][a-z0-9_]{0,63}$). Only
client capabilities are closed, because they describe the protocol rather
than your domain.
A domain of legal, juridique, regulatory or compliance triggers an
extra regime: every rule cited must carry an official source, a jurisdiction
and a verification date.
SKILL_MANIFEST_SCHEMA.json describes the shape of phases.json:
required fields, types, closed vocabularies.
The schema engine is deliberately minimal. It applies enum, minLength and
minItems, and nothing else: no pattern, no if/then, no oneOf. A
schema using those keywords would itself be rejected.
The consequence matters: conditional rules live in validator.py, which
remains the source of truth. The version rule is the example:
provides_capabilities is forbidden in a 1.0 manifest and required in a
1.1 one. That rule is enforced and tested, but it is not expressible in the
schema. Do not read required as the whole contract.
A package with only SKILL.md fails. An invalid package blocks the registry
rather than degrading silently.
phases.json.id is the identity, and SKILL.md.name must match it. The
directory must carry the same key. Keys are normalised with NFKC then
casefold, so homoglyphs cannot smuggle in a second identity. Any collision
blocks the whole build; no package is elected arbitrarily.

Every valid skill in the registry lands in exactly one category, with its reason. Nothing is discarded silently.
The only proven automatic signal is:
Platform, domain and capabilities filter only when the caller supplies those constraints. A forbidden capability rejects the skill. No semantic score is invented, and the plan is sorted by identifier.
An empty plan is explicitly valid: it carries NO_COMPATIBLE_SKILL.
The B3 plan classifies every installed skill across skills_selected,
skills_not_applicable and skills_blocked; each skill appears exactly once.
skills_missing lists capabilities with no executable provider, derived from
confirmed facts only. A gap never proves non-compliance: it says an audit
deemed necessary is not covered.
SKILL.md capped at 256 KiBCallers may only lower these limits, never raise them.
>=3.11pytest is a development dependency only.
Expected result, 778 collected, 0 failed on every platform:
The skips are platform capabilities, not failures. Linux skips the Windows-junction tests, which have no POSIX equivalent. Windows skips the two symlink tests only when the local privilege to create symlinks is missing, so a runner that holds it reports no skip at all. Every count above comes from a real run: the two Windows lines are the GitHub Actions runner and a local machine.
Normative texts are checked out with LF endings, enforced by .gitattributes.
A couple of Windows symlink tests are skipped: they need a local Windows
privilege. Windows junctions are genuinely tested.
The validator confirms one thing only:
It does not verify the real target. TARGET_VERIFIED stays forbidden in V1.
The loader refuses reparse points. Reads are bounded and confined. Output is sorted and deterministic.
One design decision deserves your attention: detect and plan take a
target path that is not confined to the configured roots, because the
point is to profile an arbitrary project. Run this server under an account
whose reach you accept, and connect it only to a trusted client. The full
threat model is in SECURITY.md.
Apache-2.0. See LICENSE and NOTICE.