claude-code-kit 0.74.0__tar.gz → 0.76.0__tar.gz

claude_code_kit-0.76.0/.claude-plugin/marketplace.json

@@ -0,0 +1,19 @@
+{
+  "name": "claude-kit",
+  "description": "claude-kit — an autonomous, stack-agnostic SDLC for Claude Code, delivered as a plugin.",
+  "owner": {
+    "name": "Arjunsingh Yadav",
+    "url": "https://github.com/ajyadav013"
+  },
+  "plugins": [
+    {
+      "name": "claude-kit",
+      "displayName": "Claude Kit SDLC",
+      "source": "./",
+      "description": "Evidence-gated SDLC for Claude Code, installed as configuration (no app code, no Docker): install CLAUDE.md + .claude/ (rules, the profile's agents/skills, hooks, artifact templates) + optional .mcp.json, then run /sdlc to drive spec → review → build → test → security → ship through profile-aware quality gates with a tamper-evident gate ledger, working memory, and a self-improving learnings loop.",
+      "version": "0.76.0",
+      "license": "MIT",
+      "keywords": ["sdlc", "agents", "orchestration", "quality-gates", "workflow", "scaffold", "cookiecutter"]
+    }
+  ]
+}

claude_code_kit-0.76.0/.claude-plugin/plugin.json

@@ -0,0 +1,25 @@
+{
+  "name": "claude-kit",
+  "displayName": "Claude Kit SDLC",
+  "version": "0.76.0",
+  "description": "Evidence-gated SDLC for Claude Code, installed as configuration (no app code, no Docker). `claude-kit init` asks ordered questions and installs CLAUDE.md + .claude/ (rules, the profile's agents/skills, hooks, artifact templates) + optional .mcp.json; run /sdlc to drive spec → review → build → test → security → ship through profile-aware quality gates with a tamper-evident gate ledger, working memory, and a self-improving learnings loop.",
+  "author": {
+    "name": "Arjunsingh Yadav",
+    "url": "https://github.com/ajyadav013"
+  },
+  "homepage": "https://github.com/ajyadav013/claude-kit",
+  "repository": "https://github.com/ajyadav013/claude-kit",
+  "license": "MIT",
+  "keywords": [
+    "sdlc",
+    "agents",
+    "orchestration",
+    "code-review",
+    "testing",
+    "quality-gates",
+    "workflow",
+    "automation",
+    "scaffold",
+    "cookiecutter"
+  ]
+}

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/CHANGELOG.md

@@ -4,6 +4,124 @@ All notable changes to claude-kit are documented here. The format follows
 [Keep a Changelog](https://keepachangelog.com/), and the project uses
 [semantic versioning](https://semver.org/).
 
+## [0.76.0] — 2026-08-01
+
+**The Trust Release — enforcement catches up with the promise.** The kit has always *claimed* that
+no gate passes without evidence; this release makes the deterministic state layer actually refuse
+the ways that claim could be quietly violated: gates closed out of order, evidence files swapped
+after the fact, snapshots corrupted mid-write, Mode E runs rejected by the very validator meant to
+protect them, and a learning-capture job that ran before anyone consented to it.
+
+### Added
+
+- **Gate ledger** (`claude-kit pipeline`): every `close-gate` now appends an entry to an append-only
+  `gate_history` in the pipeline snapshot — gate, status (`passed` / `skipped` / `overridden`),
+  evidence path **and its sha256**, a `verification` level (`agent` today; `mechanical` / `human`
+  reserved for the parser ladder in #74), and a UTC timestamp. `validate` re-hashes **every** ledger
+  entry, so tampering with an already-cited evidence file turns the run red.
+- **Gate-order enforcement.** `close-gate` derives the expected next gate from the profile's ordered
+  gate list and refuses out-of-order or already-recorded closes; `--force --override-reason` records
+  an honest `overridden` entry instead of a silent pass. The first recorded gate anchors anywhere in
+  the sequence, so adopting the ledger mid-run works.
+- **`pipeline skip-gate <gate> --reason`** — the honest path for conditional gates (e.g.
+  `contract-clear` on a stack with no API surface): recorded as `skipped` with the reason, never as
+  a fake pass.
+- **`--strict` for `pipeline validate` / `close-gate`** — a missing or unreadable install snapshot
+  fails closed instead of warning (for CI); the default stays lenient for back-compat.
+- **`claude-kit privacy-report`** — one line per installed hook: what it reads, what it writes,
+  whether it spawns a background job, whether learning capture is on, and how to turn it off.
+  Non-kit hook commands found in `settings.json` are flagged for the user's own review.
+- **Mode E** (`"mode": "E"`) is now accepted by the pipeline validator — it shipped in the
+  orchestrator in 0.58 but `pipeline.py` still rejected it. A drift-guard test now parses the mode
+  line out of `rules/continuity.md` and asserts it equals `pipeline.MODES`, so the two can't
+  diverge again.
+
+### Changed
+
+- **Learning capture is now opt-in** on every non-interactive path: the plugin manifest, the no-pip
+  starter `settings.json`, and `init --defaults` all install **no capture hooks**; `capture_mode`
+  defaults to `off`. The interactive `init` question *is* the consent — it still recommends
+  `session-end-catchup`. Recall (`load-learnings`, reading your own committed `agent-memory/`) is
+  unaffected and stays on. Documented in `SECURITY.md`.
+- **Snapshot writes are atomic and locked** — temp file + `os.replace` under an `O_EXCL` lockfile,
+  so a crash mid-`close-gate` can no longer leave a torn state file.
+- **Gates now resolve in execution order.** Fixed a real resolver bug the new order enforcement
+  uncovered: child profiles' gate lists were being union-merged behind the parent's, contradicting
+  the documented "execution-ordered" contract. `catalog/profiles.yaml` now documents the ordering
+  convention and the resolver preserves the child's declared order.
+- **Truth-in-advertising pass across every advertised surface.** The README headline now claims
+  exactly what ships ("every gate verdict must cite real command output, and the deterministic
+  state layer refuses to close a gate out of order or with unresolved findings"); "configuration
+  only" is spelled out (no app code in your repo, no daemon); the category line is now
+  **"Evidence-gated SDLC for Claude Code"** (plugin `displayName`: **Claude Kit SDLC**) across
+  plugin/marketplace manifests, PyPI metadata, `llms.txt`, and the CLI docstrings — the
+  `cookiecutter` search keywords stay. README hero restructured: proof (`examples/real-run/`)
+  before pitch, qualified install (`/plugin install claude-kit@claude-kit`), non-affiliation note
+  in the footer.
+- **Count drift fixed and pinned**: "28-agent roster" → 29 (README, `docs/agentic-patterns.md`),
+  both now anchored in `scripts/check_docs_consistency.py`; `docs/rules-context-budget.md`'s
+  numbers are annotated as the 2026-07-07 measurement snapshot; `docs/launch/posts.md` gains a
+  historical-snapshot banner.
+
+### Not adopted (deliberately)
+
+- **Splitting the plugin into core/standard/enterprise packages.** Profiles already gate what a
+  given install receives; a package split would break the one-source-of-truth layout for a
+  distribution problem no user has reported. Revisit with adoption data.
+- **A config-file source of truth for pipeline modes.** A drift-guard test asserting
+  `rules/continuity.md` ≡ `pipeline.MODES` is cheaper than a third artifact both would then have to
+  agree with.
+- **Mechanical evidence parsing in this release** (JUnit/coverage/SARIF verdicts refusing a close).
+  Deliberately sequenced as the next release so this one stays reviewable; the ledger's
+  `verification` enum already reserves `mechanical` / `human` for it (issue #74).
+
+## [0.75.0] — 2026-07-31
+
+**The Devil's Advocate learns to weigh, not only to prosecute — a premortem, a merits-and-costs
+balance sheet, and a third verdict that records an accepted trade-off instead of swallowing it.**
+The agent already gated the two moments where consensus is most dangerous (the plan before EM
+approval, and any gate reached by *unanimous* PASS). What it could not do was say "this approach is
+sound, and here is what it costs" — its output was findings plus a binary verdict, so a known
+trade-off either became a blocking finding or vanished into a clean `CONFIRMED`.
+
+### Changed
+
+- **`agents/devils-advocate.md`** gains two steps and one verdict:
+  - **Premortem** — findings attack the artifact *as written*; the premortem attacks it *as it will be
+    operated*. "Assume this shipped and failed — the likeliest cause, and the earliest signal that
+    would have caught it." At the plan gate this is the highest-value moment to run one, since nothing
+    has been built yet.
+  - **Balance sheet** — merits alongside costs. A critique listing only defects cannot be weighed
+    against doing nothing, and leaves no record of *why* a downside was accepted.
+  - **`CONFIRMED-WITH-COSTS`** — passes the gate while naming a durable cost, each with an accepting
+    role and a concrete revisit trigger, recorded in `CONTINUITY.md`.
+- **`rules/quality-gates.md`** · **`rules/mandatory-workflow.md`** (§1e.5, §3b.5) ·
+  **`agents/orchestrator.md`** (Stage PC, test-coverage gate) — verdict vocabulary and gate criteria
+  updated in step; a gate now clears on `CONFIRMED` *or* `CONFIRMED-WITH-COSTS`.
+
+### Guarded against
+
+- **Gate erosion.** A third "pass, but…" verdict is exactly the kind of state that quietly becomes the
+  default, so `CONFIRMED-WITH-COSTS` is deliberately *more* expensive to produce than a plain
+  `CONFIRMED`: it can never carry a Critical/High/Medium (anything blocking is still `UPHELD`), and a
+  cost that cannot name an owner and a revisit trigger is declared a hedge and dropped.
+- **A loophole found by dogfooding this change.** Reviewing the diff under the kit's own discipline:
+  rule 2 read *"you must do real work before CONFIRMED"* — pinned to that literal token, leaving the
+  new verdict outside the anti-sycophancy requirement and rubber-stampable with zero probes. Now
+  *"before any confirming verdict."* Precisely the class of miss the agent exists to catch.
+
+### Not adopted (deliberately)
+
+- **A separate `critic-reviewer` agent.** It would duplicate `devils-advocate` (reuse-first, golden
+  rule #8) and drag in a catalog entry, a docs row, count updates and test changes. Extending the
+  existing agent changed no counts at all.
+- **Per-story adversarial critique** (a pass on each story's chosen approach between breakdown and
+  implementation). It closes a real gap — story-level *approach* decisions are reviewed for
+  correctness but never adversarially — at the cost of one opus spawn per story. Deferred, not
+  rejected; revisit if approach-level mistakes are observed slipping past the two existing gates.
+- **A risk-gated variant of the above** (fire only for `risk-classifier` medium+ or one-way-door
+  stories). The cheaper shape of the same idea, held back with it.
+
 ## [0.74.0] — 2026-07-28
 
 **The kit learns to build AI products, not just to be one — 63 attributed digests from a 72-item

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/CLAUDE.md

@@ -4,8 +4,9 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 
 # Developing claude-kit
 
-This repository **is** claude-kit — a Cookiecutter-style scaffolder that installs a stack-agnostic,
-autonomous-SDLC **configuration** (no application code, no Docker) into a Claude Code project. It is
+This repository **is** claude-kit — an evidence-gated SDLC for Claude Code: a scaffolder that
+installs a stack-agnostic, autonomous-SDLC **configuration** (no application code, no Docker) into
+a Claude Code project. It is
 distributed two ways from one source of truth:
 
 1. **As a Claude Code plugin** — components are auto-discovered from the repo root.
@@ -122,7 +123,7 @@ root are read directly by the plugin **and** bundled into the wheel (mapped to
 
 ## Dogfooding / local testing
 
-- **Plugin:** `claude` → `/plugin marketplace add .` → `/plugin install claude-kit` (loads the
+- **Plugin:** `claude` → `/plugin marketplace add .` → `/plugin install claude-kit@claude-kit` (loads the
   agents/skills/commands/hooks from this checkout).
 - **CLI:** `pip install -e '.[dev]'` then `claude-kit init /tmp/demo --defaults` (or interactive),
   `claude-kit validate /tmp/demo`, `claude-kit diff /tmp/demo`, and inspect the result.

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/CONTRIBUTING.md

@@ -59,7 +59,7 @@ locally, and how to release.
 
 ```bash
 # Plugin (dogfood this checkout):
-#   in Claude Code:  /plugin marketplace add .   then   /plugin install claude-kit
+#   in Claude Code:  /plugin marketplace add .   then   /plugin install claude-kit@claude-kit
 
 # CLI:
 pip install -e '.[dev]'
@@ -107,6 +107,9 @@ a specific stack — `pytest` enforces the no-Docker invariant on a scaffolded p
 4. CI auto-publishes to PyPI on merge to `main` when the version is new (`.github/workflows/publish.yml`,
    OIDC trusted publishing). Manual `python3 -m twine upload dist/*` is the fallback.
 5. Tag the release and push so plugin users get the update.
+6. Publish the context-cost report for the release: run `claude plugin details claude-kit@claude-kit`
+   in Claude Code and paste the token/context figures into the release notes, so users can see what
+   the plugin costs their context window before installing.
 
 ## License
 

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: claude-code-kit
-Version: 0.74.0
-Summary: Cookiecutter-style scaffolder for an autonomous Claude Code SDLC configuration (no app code, no Docker). Asks ordered questions and installs CLAUDE.md + .claude/ (rules, the chosen profile's agents/skills, hooks, artifact templates) + optional .mcp.json; run /sdlc to drive spec → review → build → test → security → ship through profile-aware quality gates, working memory, and a self-improving learnings loop.
+Version: 0.76.0
+Summary: Evidence-gated SDLC for Claude Code, installed as configuration (no app code, no Docker). Asks ordered questions and installs CLAUDE.md + .claude/ (rules, the chosen profile's agents/skills, hooks, artifact templates) + optional .mcp.json; run /sdlc to drive spec → review → build → test → security → ship through profile-aware quality gates with a tamper-evident gate ledger, working memory, and a self-improving learnings loop.
 Project-URL: Homepage, https://github.com/ajyadav013/claude-kit
 Project-URL: Repository, https://github.com/ajyadav013/claude-kit
 Project-URL: Issues, https://github.com/ajyadav013/claude-kit/issues
@@ -35,13 +35,29 @@ Description-Content-Type: text/markdown
 
 # claude-kit
 
+### Evidence-gated SDLC for Claude Code
+
 **Turn Claude Code into a disciplined engineering team: one command — `/sdlc <task>` — runs your
 request through spec → review → code → test → security → PR, with a quality gate between every
 phase.**
 
-The differentiator is trust: **no gate passes without real, cited command output**, so a green
-result means the checks actually ran. It installs as **configuration only** — no runtime, no
-application code, no Docker.
+The differentiator is trust: **every gate verdict must cite real command output, and the
+deterministic state layer refuses to close a gate out of order or with unresolved
+Critical/High/Medium findings.** It installs as **configuration, not a runtime** — no application
+code in your repo and no daemon: Claude Code configuration, local hooks, and an optional CLI.
+
+</div>
+
+**Proof over pitch:** in the captured run in [`examples/real-run/`](examples/real-run/), a unanimous
+review PASS triggered the adversarial `devils-advocate` — it caught a Medium bug every reviewer had
+missed, and the gate refused to advance until the fix landed. The run is checked in verbatim: state
+file, agent verdicts, diff, asciicast.
+
+```text
+/plugin marketplace add ajyadav013/claude-kit
+/plugin install claude-kit@claude-kit
+/claude-kit:sdlc Add a /health endpoint that returns the build version
+```
 
 <!-- DEMO PLACEHOLDER — a 60–90s terminal capture of a gated `/sdlc` run belongs here once recorded.
      Shot list, timing, and which real assets to show: docs/launch/demo-script.md.
@@ -49,6 +65,8 @@ application code, no Docker.
      Medium bug a unanimous review missed, and the gate refusing to advance until it was fixed).
      Until the GIF exists, that folder IS the evidence — linked from "How it works" and "The pipeline" below. -->
 
+<div align="center">
+
 [![PyPI](https://img.shields.io/pypi/v/claude-code-kit.svg)](https://pypi.org/project/claude-code-kit/)
 [![Python](https://img.shields.io/pypi/pyversions/claude-code-kit.svg)](https://pypi.org/project/claude-code-kit/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
@@ -89,16 +107,13 @@ The fastest path to a gated run — **no CLI install, no `init`, no restart:**
 
 ```text
 /plugin marketplace add ajyadav013/claude-kit
-/plugin install claude-kit
+/plugin install claude-kit@claude-kit        # qualified name: the claude-kit plugin from the claude-kit marketplace
 /claude-kit:sdlc Add a /health endpoint that returns the build version
 ```
 
 That's the whole loop in three lines: the standard pipeline runs immediately — spec → review →
-build → test, gate by gate — with every verdict backed by real output.
-
-**Proof it works:** [`examples/real-run/`](examples/real-run/) is a genuine captured run where the
-`devils-advocate` caught a Medium bug a unanimous review missed, and the gate refused to advance
-until it was fixed.
+build → test, gate by gate — with every verdict backed by real output. (The proof run lives in
+[`examples/real-run/`](examples/real-run/) — see the top of this page.)
 
 When you want the pipeline tuned to *your* repo — your stack, commands, rigor profile, and the
 safety hooks — run `init`:
@@ -373,7 +388,7 @@ Report vulnerabilities privately — see [`SECURITY.md`](SECURITY.md).
 |---|---|
 | [`docs/install.md`](docs/install.md) | Every install detail: prerequisites, Windows, plugin updates, all `init` questions, `init.yaml`, what lands on disk |
 | [`docs/cli.md`](docs/cli.md) | Full CLI command reference, safe-upgrade mechanics, troubleshooting |
-| [`docs/agents.md`](docs/agents.md) | How to drive the agents + the full 28-agent roster and per-run cost |
+| [`docs/agents.md`](docs/agents.md) | How to drive the agents + the full 29-agent roster and per-run cost |
 | [`docs/architecture.md`](docs/architecture.md) | Diagrams: distribution, catalog resolution, the state machine — and how to extend via the catalog |
 | [`docs/influences.md`](docs/influences.md) | The reuse-first adoption history: what we learned, shipped, and deliberately skipped |
 | [`docs/autonomous-operation.md`](docs/autonomous-operation.md) | Unattended runs: permission modes × autonomy levels, headless mode, the bounded loop script, CI-trigger design |
@@ -391,10 +406,12 @@ Report vulnerabilities privately — see [`SECURITY.md`](SECURITY.md).
 Issues and PRs welcome — see [`CONTRIBUTING.md`](CONTRIBUTING.md). To dogfood a local checkout:
 
 ```bash
-# As a plugin:  /plugin marketplace add .   then   /plugin install claude-kit
+# As a plugin:  /plugin marketplace add .   then   /plugin install claude-kit@claude-kit
 # As the CLI:   pip install -e '.[dev]'   then   claude-kit init /tmp/demo --defaults   &&   pytest
 ```
 
 ## License
 
 [MIT](LICENSE) © Arjunsingh Yadav
+
+claude-kit is an independent open-source project — not affiliated with or endorsed by Anthropic.

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/README.md

@@ -2,13 +2,29 @@
 
 # claude-kit
 
+### Evidence-gated SDLC for Claude Code
+
 **Turn Claude Code into a disciplined engineering team: one command — `/sdlc <task>` — runs your
 request through spec → review → code → test → security → PR, with a quality gate between every
 phase.**
 
-The differentiator is trust: **no gate passes without real, cited command output**, so a green
-result means the checks actually ran. It installs as **configuration only** — no runtime, no
-application code, no Docker.
+The differentiator is trust: **every gate verdict must cite real command output, and the
+deterministic state layer refuses to close a gate out of order or with unresolved
+Critical/High/Medium findings.** It installs as **configuration, not a runtime** — no application
+code in your repo and no daemon: Claude Code configuration, local hooks, and an optional CLI.
+
+</div>
+
+**Proof over pitch:** in the captured run in [`examples/real-run/`](examples/real-run/), a unanimous
+review PASS triggered the adversarial `devils-advocate` — it caught a Medium bug every reviewer had
+missed, and the gate refused to advance until the fix landed. The run is checked in verbatim: state
+file, agent verdicts, diff, asciicast.
+
+```text
+/plugin marketplace add ajyadav013/claude-kit
+/plugin install claude-kit@claude-kit
+/claude-kit:sdlc Add a /health endpoint that returns the build version
+```
 
 <!-- DEMO PLACEHOLDER — a 60–90s terminal capture of a gated `/sdlc` run belongs here once recorded.
      Shot list, timing, and which real assets to show: docs/launch/demo-script.md.
@@ -16,6 +32,8 @@ application code, no Docker.
      Medium bug a unanimous review missed, and the gate refusing to advance until it was fixed).
      Until the GIF exists, that folder IS the evidence — linked from "How it works" and "The pipeline" below. -->
 
+<div align="center">
+
 [![PyPI](https://img.shields.io/pypi/v/claude-code-kit.svg)](https://pypi.org/project/claude-code-kit/)
 [![Python](https://img.shields.io/pypi/pyversions/claude-code-kit.svg)](https://pypi.org/project/claude-code-kit/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
@@ -56,16 +74,13 @@ The fastest path to a gated run — **no CLI install, no `init`, no restart:**
 
 ```text
 /plugin marketplace add ajyadav013/claude-kit
-/plugin install claude-kit
+/plugin install claude-kit@claude-kit        # qualified name: the claude-kit plugin from the claude-kit marketplace
 /claude-kit:sdlc Add a /health endpoint that returns the build version
 ```
 
 That's the whole loop in three lines: the standard pipeline runs immediately — spec → review →
-build → test, gate by gate — with every verdict backed by real output.
-
-**Proof it works:** [`examples/real-run/`](examples/real-run/) is a genuine captured run where the
-`devils-advocate` caught a Medium bug a unanimous review missed, and the gate refused to advance
-until it was fixed.
+build → test, gate by gate — with every verdict backed by real output. (The proof run lives in
+[`examples/real-run/`](examples/real-run/) — see the top of this page.)
 
 When you want the pipeline tuned to *your* repo — your stack, commands, rigor profile, and the
 safety hooks — run `init`:
@@ -340,7 +355,7 @@ Report vulnerabilities privately — see [`SECURITY.md`](SECURITY.md).
 |---|---|
 | [`docs/install.md`](docs/install.md) | Every install detail: prerequisites, Windows, plugin updates, all `init` questions, `init.yaml`, what lands on disk |
 | [`docs/cli.md`](docs/cli.md) | Full CLI command reference, safe-upgrade mechanics, troubleshooting |
-| [`docs/agents.md`](docs/agents.md) | How to drive the agents + the full 28-agent roster and per-run cost |
+| [`docs/agents.md`](docs/agents.md) | How to drive the agents + the full 29-agent roster and per-run cost |
 | [`docs/architecture.md`](docs/architecture.md) | Diagrams: distribution, catalog resolution, the state machine — and how to extend via the catalog |
 | [`docs/influences.md`](docs/influences.md) | The reuse-first adoption history: what we learned, shipped, and deliberately skipped |
 | [`docs/autonomous-operation.md`](docs/autonomous-operation.md) | Unattended runs: permission modes × autonomy levels, headless mode, the bounded loop script, CI-trigger design |
@@ -358,10 +373,12 @@ Report vulnerabilities privately — see [`SECURITY.md`](SECURITY.md).
 Issues and PRs welcome — see [`CONTRIBUTING.md`](CONTRIBUTING.md). To dogfood a local checkout:
 
 ```bash
-# As a plugin:  /plugin marketplace add .   then   /plugin install claude-kit
+# As a plugin:  /plugin marketplace add .   then   /plugin install claude-kit@claude-kit
 # As the CLI:   pip install -e '.[dev]'   then   claude-kit init /tmp/demo --defaults   &&   pytest
 ```
 
 ## License
 
 [MIT](LICENSE) © Arjunsingh Yadav
+
+claude-kit is an independent open-source project — not affiliated with or endorsed by Anthropic.

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/agents/devils-advocate.md

@@ -1,6 +1,6 @@
 ---
 name: devils-advocate
-description: Anti-sycophancy adversarial reviewer. Critiques a plan/spec before approval, and when a gate reaches unanimous PASS. Gates the pipeline — the artifact is not final until it returns CONFIRMED.
+description: Anti-sycophancy adversarial reviewer. Critiques a plan/spec before approval, and when a gate reaches unanimous PASS. Returns findings, a premortem, and a merits-and-costs balance sheet. Gates the pipeline — nothing is final until it confirms.
 tools: Read, Glob, Grep, Bash, SendMessage
 permissionMode: plan
 model: opus
@@ -39,6 +39,8 @@ You are spawned by the Orchestrator at **two moments**, both chosen because cons
    - Error/empty/loading states actually wired, not just the success path.
    - Spec drift — undocumented behavior added, or a requirement quietly dropped.
 4. **Prove it.** Where feasible, demonstrate the issue (a failing scenario, a grep showing the missing filter, a file:line reference). A claim without evidence is not a finding.
+5. **Run a premortem.** Findings attack the artifact as written; the premortem attacks it as it will be *operated*. Assume this shipped and failed six months from now — state the single most likely cause, and the earliest signal that would have caught it. A failure mode with no owning finding is itself a finding when it is Critical/High/Medium; when it is genuinely a cost rather than a defect, it belongs in the balance sheet below.
+6. **State the merits, not only the demerits.** A critique that lists only what is wrong cannot be weighed against the alternative of doing nothing, and gives the reader no way to judge whether an accepted cost is worth paying. Name what the approach gets right and what it demonstrably costs, so the decision is legible later.
 
 ## Output
 
@@ -52,16 +54,26 @@ Effort: {what you specifically probed that they did not}
 1. [Critical|High|Medium|Low] `{file}:{line}` — {issue} — {evidence/repro}
    (or: "No blocking issue found in {area} — checked {what}")
 
-## Verdict: {UPHELD | CONFIRMED}
+## Premortem
+Assume this shipped and failed. Most likely cause: {failure mode}.
+Earliest signal: {what would have shown it first — a metric, a log, a test}.
+
+## Balance sheet
+Merits: {what this approach genuinely gets right, and what it buys}
+Costs:  {what it gives up — named, not hedged; "none" is a claim you must defend}
+
+## Verdict: {UPHELD | CONFIRMED-WITH-COSTS | CONFIRMED}
 - UPHELD  -> at least one Critical/High/Medium found. Gate FAILS. Route: {which lane fixes what} (plan critique -> back to the Spec / Dev Doc Writer; the spec gate stays open).
-- CONFIRMED -> genuinely clean after adversarial review. Gate may PASS.
+- CONFIRMED-WITH-COSTS -> no blocking finding, but the balance sheet carries a cost that outlives this gate. Gate PASSes. For each cost: {the cost} | accepted by {role} | revisit when {concrete trigger}.
+- CONFIRMED -> genuinely clean after adversarial review, no outstanding cost worth recording. Gate may PASS.
 ```
 
 ## Rules
 
 1. **You do not write code.** You probe, prove, and rule.
-2. **You must do real work before CONFIRMED.** "Looks fine" is not allowed — name what you attacked and why it held. A CONFIRMED with no described probes is itself a failure.
+2. **You must do real work before any confirming verdict.** "Looks fine" is not allowed — name what you attacked and why it held. A CONFIRMED or CONFIRMED-WITH-COSTS with no described probes is itself a failure.
 3. **Classify by the shared severity model.** Only Critical/High/Medium block; Low/Cosmetic are notes.
 4. **No sycophancy, no nihilism.** Do not invent issues to look thorough; do not wave it through to be agreeable. Report what is actually there.
-5. **One pass.** You run once per plan critique and once per unanimous gate. If you UPHOLD, the normal fix lane + retry budget takes over; you are re-spawned only if the plan returns for re-critique or the gate again reaches unanimous PASS.
-6. Include any blind-spot pattern you find (e.g., "all reviewers missed tenant filter on list endpoints") in your verdict message — you run read-only, so the Orchestrator records it in `CONTINUITY.md` (and promotes it to `agent-memory/` if it is a recurring class of miss) on your behalf.
+5. **CONFIRMED-WITH-COSTS is not a softer UPHELD.** It never carries a Critical/High/Medium — anything blocking is UPHELD, full stop. Use it only when the artifact is correct *and* you can name a durable cost with an owner and a concrete revisit trigger. A cost you cannot attribute or cannot say when to revisit is not a cost, it is a hedge: drop it and return CONFIRMED.
+6. **One pass.** You run once per plan critique and once per unanimous gate. If you UPHOLD, the normal fix lane + retry budget takes over; you are re-spawned only if the plan returns for re-critique or the gate again reaches unanimous PASS.
+7. Include any blind-spot pattern you find (e.g., "all reviewers missed tenant filter on list endpoints") in your verdict message — you run read-only, so the Orchestrator records it in `CONTINUITY.md` (and promotes it to `agent-memory/` if it is a recurring class of miss) on your behalf.

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/agents/orchestrator.md

@@ -315,8 +315,12 @@ Before treating the review chain's approval as final, run an adversarial pass on
 - **Spawn**: `devils-advocate` with the spec + developer documentation (and the review-chain verdicts).
 - It argues the plan is wrong — weakest/most-volatile requirement, untestable acceptance criterion,
   hidden dependency, missing requirement, unjustified scope, the step most likely to fail.
-- **Gate**: a **CONFIRMED** verdict lets the Story Planner proceed; an **UPHELD** verdict (any
-  Critical/High/Medium) routes back to the **spec-doc-writer** and the spec gate stays open.
+- It also returns a **premortem** and a **merits-and-costs balance sheet** — record both with the gate
+  evidence, so the plan's accepted trade-offs are legible when the work is reviewed later.
+- **Gate**: a **CONFIRMED** verdict lets the Story Planner proceed; so does **CONFIRMED-WITH-COSTS**,
+  whose named costs you write to `CONTINUITY.md` (each with its accepting role and revisit trigger)
+  before proceeding. An **UPHELD** verdict (any Critical/High/Medium) routes back to the
+  **spec-doc-writer** and the spec gate stays open.
 - **Profile**: standard and enterprise only — `devils-advocate` isn't installed in **lean**, where the
   spec-doc-writer's own self-critique (its RARV cycle) is the safeguard. Skip with a noted reason in lean.
 
@@ -483,7 +487,7 @@ For full-stack work or features with significant scope, **spawn multiple testers
   - All defects have clear classification (API / UI / integration)
   - All defects have reproduction steps
 - **Blind review**: the three senior testers assess **independently** — none sees another's findings — and each returns PASS/FAIL with severity-classified findings. Any Critical/High/Medium → gate FAILs.
-- **Devil's Advocate (anti-sycophancy)**: if all three return a **unanimous PASS**, **spawn `devils-advocate`** before the gate may pass. It assumes the work is guilty and hunts for what everyone missed. VERIFIED requires its CONFIRMED verdict; an UPHELD verdict re-opens the Defect Loop. See `.claude/rules/quality-gates.md`.
+- **Devil's Advocate (anti-sycophancy)**: if all three return a **unanimous PASS**, **spawn `devils-advocate`** before the gate may pass. It assumes the work is guilty and hunts for what everyone missed. VERIFIED requires a CONFIRMED or CONFIRMED-WITH-COSTS verdict (record the named costs in `CONTINUITY.md`); an UPHELD verdict re-opens the Defect Loop. See `.claude/rules/quality-gates.md`.
 - **Gate**: `VERIFIED` from merge-reviewer (plus CONFIRMED from `devils-advocate` when the senior testers were unanimous).
 - On FAIL from any tester or senior tester → enter **Defect Loop**.
 
@@ -827,7 +831,7 @@ silently, and never marked PASS. Every stage that *is* active is mandatory.
 14. **Verify outputs exist.** Check that expected files are created before marking a stage complete.
 15. **Prefer parallel over sequential.** If two stages have no data dependency, run them in parallel.
 16. **Persist working memory.** Read/write `.claude/CONTINUITY.md` every turn and at every stage transition; recover from it after compaction. Mirror gate-precise state into `.claude/state/pipeline-snapshot.json` and resume from it by *reloading* (re-enter after `last_gate_passed`), never by re-running passed gates or re-applying committed edits.
-17. **Anti-sycophancy.** In standard+, the plan is critiqued by `devils-advocate` before approval is final (Stage PC); and a unanimous PASS at the test-coverage gate is not VERIFIED until `devils-advocate` returns CONFIRMED.
+17. **Anti-sycophancy.** In standard+, the plan is critiqued by `devils-advocate` before approval is final (Stage PC); and a unanimous PASS at the test-coverage gate is not VERIFIED until `devils-advocate` returns CONFIRMED or CONFIRMED-WITH-COSTS. Every verdict carries a premortem and a merits-and-costs balance sheet; record accepted costs in `CONTINUITY.md`.
 18. **Operability gates.** For deployable/observable changes, run DevOps (Pipeline Green) and Observability (Observability Ready) before the PR Raiser.
 19. **Name the skills in every spawn.** Each worker prompt states which skill(s) to load for its stage (Skill Routing table); never assume a worker will find them itself.
 20. **Disjoint boundaries in every parallel spawn.** Every parallel worker prompt names its exact file boundary; concurrent boundaries never overlap.

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/catalog/capture.yaml

@@ -15,9 +15,13 @@
 
 version: 1
 
-# Used by `--defaults`, the plugin install, and shown as the recommended pick. The interactive prompt
-# overrides this to `off` for the lean profile (lean is intentionally minimal).
-default: session-end-catchup
+# CONSENT GATE (0.76.0): background capture reads session transcript content, so it is opt-in —
+# `default` (used by every NON-interactive path: `--defaults`, config files without the key) is
+# `off`. The interactive `init` prompt asks explicitly and preselects `recommended` (off for the
+# lean profile — lean is intentionally minimal). The plugin channel and the no-pip starter ship no
+# capture hooks at all (see PLUGIN_HOOK_IDS / STARTER_HOOK_IDS in src/claude_kit/hooks.py).
+default: "off"
+recommended: session-end-catchup
 
 modes:
   "off":

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/catalog/profiles.yaml

@@ -5,6 +5,12 @@
 # another profile (this profile's lists are unioned with the inherited ones). The special value
 # `all` means "every available item discovered in the payload" (agents/, skills/, hooks/).
 #
+# ORDER CONVENTION for `gates:` — gates are the one list where order is data: the resolved list
+# lands execution-ordered in stack-catalog.snapshot.yaml and `claude-kit pipeline close-gate`
+# enforces it. A profile that inherits MUST therefore declare its FULL gate list in execution
+# order (standard and enterprise below do); the resolver keeps the profile's own order and lets
+# inherited gates only fill gaps.
+#
 # Agent and skill names must match files in agents/ and skills/. Hook ids map to entries in the
 # hook registry in scaffold.py (script-backed hooks copy their .sh; inline/prompt hooks don't).
 

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/agentic-patterns.md

@@ -23,9 +23,9 @@ they are already embodied in its rules, agents, skills, and the orchestration mo
 | 4 | Reflection | `rules/rarv-cycle.md` (Reflect); `agents/devils-advocate.md`; `skills/doubt-driven-development`; blind review in `rules/quality-gates.md` | covered (strong) |
 | 5 | Tool Use | each agent's `tools:` frontmatter; MCP integration; design principles in **`rules/tool-design.md`** (new) | covered (rule added) |
 | 6 | Planning | `agents/spec-doc-writer`, `story-planner`, `technical-architect`; `skills/spec-driven-development`, `planning-and-task-breakdown`, `scope`, `sprint` | covered (strong) |
-| 7 | Multi-Agent Collaboration | 28 agents; subagents + Agent-Teams worked example in orchestration-patterns | covered (strong) |
+| 7 | Multi-Agent Collaboration | 29 agents; subagents + Agent-Teams worked example in orchestration-patterns | covered (strong) |
 | 8 | Memory Management | `rules/continuity.md` (working) + `rules/agent-memory.md` (durable, now with the **working/episodic/semantic/procedural** taxonomy); `skills/remember`, `consolidate-learnings`, `context-engineering` | covered (taxonomy added) |
-| 9 | Learning & Adaptation | `rules/agent-memory.md`; `skills/remember`; the `load-learnings` SessionStart hook; **auto-capture** — the configurable `capture-learnings` hook (a non-blocking background job over Claude's own work, fired per the init-time `capture_mode`: SessionEnd · +SessionStart catch-up for abrupt closes · per-task Stop) | covered |
+| 9 | Learning & Adaptation | `rules/agent-memory.md`; `skills/remember`; the `load-learnings` SessionStart hook; **auto-capture** — the **opt-in** `capture-learnings` hook (a non-blocking background job over Claude's own work, fired per the init-time `capture_mode`, default off: SessionEnd · +SessionStart catch-up for abrupt closes · per-task Stop) | covered |
 | 10 | Model Context Protocol (MCP) | `catalog/mcp.yaml` → `.mcp.json`; `skills/browser-testing-with-devtools`; design guidance in `rules/tool-design.md` | covered (strong) |
 | 11 | Goal Setting & Monitoring | **`rules/goal-setting-and-monitoring.md`** (new) + spec acceptance criteria + `acceptance-reviewer` + quality-gates process signals | **rule added** |
 | 12 | Exception Handling & Recovery | **`rules/agent-resilience.md`** (new) + gate retry budgets + `skills/debugging-and-error-recovery` (product bugs) | **rule added** |

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/agents.md

@@ -4,7 +4,7 @@ claude-kit installs a team of focused agents plus an **orchestrator** that runs
 software-delivery pipeline with a quality gate between phases. This guide shows how to drive them.
 
 > Prerequisite: the agents load when claude-kit is active in a project — installed as a plugin
-> (`/plugin install claude-kit`) or scaffolded with `claude-kit init`. Which agents are present
+> (`/plugin install claude-kit@claude-kit`) or scaffolded with `claude-kit init`. Which agents are present
 > depends on the **profile** you chose at init (`lean ⊊ standard ⊊ enterprise`). After installing
 > into a project, **restart Claude Code** so the agents, skills, and hooks load.
 >

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/architecture.md

@@ -26,7 +26,7 @@ flowchart LR
     SRC -->|"auto-discovered at repo root"| MKT["GitHub marketplace<br/>.claude-plugin/marketplace.json"]
 
     PKG -->|"pip install claude-code-kit<br/>claude-kit init"| PROJ
-    MKT -->|"/plugin marketplace add ajyadav013/claude-kit<br/>/plugin install claude-kit"| CC["Claude Code session<br/>(agents · skills · commands · hooks live)"]
+    MKT -->|"/plugin marketplace add ajyadav013/claude-kit<br/>/plugin install claude-kit@claude-kit"| CC["Claude Code session<br/>(agents · skills · commands · hooks live)"]
     CC -->|"/claude-kit:init"| PROJ["Your project:<br/>CLAUDE.md + .claude/{rules,agents,skills,hooks,templates,config}"]
 
     PROJ --> RUN(["/sdlc — autonomous SDLC active"])

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/cli.md

@@ -14,8 +14,9 @@ exports the kit's configuration. Install: `pip install claude-code-kit` (see
 | `diff [path]` | Preview what an `upgrade` would change (no writes) |
 | `export [path] -t cursor\|agents\|copilot [--force] [--dry-run] [--json]` | Project the config into Cursor (`.cursor/`), a root `AGENTS.md`, or GitHub Copilot (`.github/copilot-instructions.md`) for editors that aren't Claude Code |
 | `upgrade [path] [--force]` | Refresh kit/overlay files; protect your edits; prune orphans |
-| `pipeline validate · status · close-gate · abort` | Inspect/mutate the `/sdlc` state files (gate/lane/evidence coherence); **does not run** the pipeline |
+| `pipeline validate · status · close-gate · skip-gate · abort` | Inspect/mutate the `/sdlc` state files; **does not run** the pipeline. `close-gate` appends to the gate ledger (evidence sha256 + UTC timestamp) and refuses out-of-order or findings-blocked closes; `skip-gate --reason` records a conditional gate that doesn't apply; `validate` re-hashes **every** ledger entry; `--strict` fails closed when the install snapshot is missing (CI) |
 | `list-options` | List available frontend/backend/database/profile/MCP options |
+| `privacy-report [path] [--json]` | One line per installed hook: what it reads/writes/spawns; flags non-kit hook commands; states whether background learning capture is on and how to disable it |
 | `status [path]` | Show what's installed, the selection, and working memory |
 | `tickets [ID] [--path DIR] [--graph\|--graph-git] [--html] [--watch N] [--json]` | Live ticket board with tokens / model / agent / elapsed per ticket; `--graph` for the dependency graph, `--graph-git` for the commit graph, `--html` for a browser Kanban board, an `ID` for the detail view |
 | `version` | Print the version |

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/install.md

@@ -22,7 +22,7 @@ Makes all agents, skills, commands, and hooks available inside Claude Code:
 
 ```text
 /plugin marketplace add ajyadav013/claude-kit
-/plugin install claude-kit
+/plugin install claude-kit@claude-kit        # qualified name: the claude-kit plugin from the claude-kit marketplace
 ```
 
 Then, inside any project you want the pipeline to manage:

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/launch/github-about.md

@@ -9,7 +9,7 @@ Internal checklist for setting the GitHub repository "About" sidebar (descriptio
 Ready-to-paste candidates (kept within a self-imposed 120-character budget):
 
 1. **Recommended:**  
-   `Stack-agnostic SDLC config scaffolder for Claude Code. Quality gates, 28 agents, no Docker. MIT. pip install.` (109 chars)
+   `Evidence-gated SDLC for Claude Code — gates close only on cited evidence. Stack-agnostic config, no Docker. MIT.` (113 chars)
 
 2. **Alternative (simpler):**  
    `Autonomous SDLC pipeline for Claude Code. Spec→review→build→test→ship. Stack-agnostic config, no app code.` (106 chars)

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/launch/posts.md

@@ -1,5 +1,9 @@
 # Launch Announcement Drafts
 
+> **Historical snapshot.** These drafts were written for an earlier release and are kept for
+> reference — component counts and claims reflect the kit at that time. The README carries the
+> current numbers (enforced by `scripts/check_docs_consistency.py`).
+
 ## Show HN
 
 ### Title

{claude_code_kit-0.74.0 → claude_code_kit-0.76.0}/docs/rules-context-budget.md

@@ -22,7 +22,8 @@ exactly: OK"` runs, 2026-07-07; total input = `input + cache_creation + cache_re
 | Same project, `.claude/rules/` removed | 66,728 |
 
 **Attribution: `.claude/rules/` = 97,031 tokens (93% of the kit's 103,916-token injection).
-Everything else the kit installs — CLAUDE.md, 42 skills, 28 agents, hooks, settings — totals
+Everything else the kit installs — CLAUDE.md, 42 skills, 28 agents, hooks, settings (component
+counts as of the 2026-07-07 measurement; a historical snapshot, not the live roster) — totals
 6,885 tokens.** The lean design works everywhere except the one directory the original design
 assumed was inert, on-demand reference material. `CLAUDE.md`'s "live on-demand in
 `.claude/rules/`" sentence is false under real semantics.