atomadic-forge 0.83.0__tar.gz → 0.87.0__tar.gz

{atomadic_forge-0.83.0 → atomadic_forge-0.87.0}/.github/actions/forge-action/README.md

@@ -104,7 +104,7 @@ That's it. The action will:
 
 ╔══════════════════════════════════════════════════════════╗
 │ Atomadic Forge Receipt         atomadic-forge.receipt/v1 │
-│ ✓ PASS                                       forge 0.2.2 │
+│ ✓ PASS                                       forge x.y.z │
 │ atomadic-forge                                           │
 │ CERTIFY                                      100.0 / 100 │
 │   docs ✓  tests ✓  layout ✓  wire ✓                      │

{atomadic_forge-0.83.0 → atomadic_forge-0.87.0}/CHANGELOG.md

@@ -2,15 +2,46 @@
 
 ## Unreleased
 
+## v0.87.0 — 2026-05-10
+
+- feat(context-pack): add a deterministic, token-budgeted `repo_map` and shared `repo_state` capsule so first-call agent context stays anchored without broad rescans
+- feat(recon): add per-file recon-cache manifests and incremental merge logic so changed, removed, and rename-only file edits do not force full scout walks or return stale paths
+- feat(welcome): reuse fresh scout artifacts when possible, default to structural-first certify for speed, and route non-PASS certify outcomes back to full `forge certify . --json`
+- fix(docs): refresh registry metadata, agent docs, release guides, and usage docs so the published Forge surface matches 11 tools, 92 actions, 10 recipes, and the current release flow
+
+## v0.86.0 — 2026-05-09
+
+- feat(mcp): add `inspect` tool — 10 schema-validated AST inspection actions split from explore's undocumented catch-all dispatch (body, symbol, class_signals, ast_scope_blueprint, mint_ast_locks, add_node, classify_repo_layout, fingerprint_repo, synthesis_plan, swarm_repo_specs)
+- fix(metrics): remove `max()` inflation guard in `forge_metrics_feature.py` — mcp_action_count now reflects the live schema count (92) instead of holding an inflated historical value (116)
+- chore(release): bump to v0.86.0; update forge_metrics.json (11 tools, 92 actions); update MCP_TOOLS.md and module docstring
+
+## v0.85.1 — 2026-05-09
+
+- fix(certify): bump `run_pytest` default timeout from 60 s → 180 s so the full 1635-test suite completes on Windows without hitting the timeout gate
+- fix(metrics): regenerate `forge_metrics.json` to v0.85.0 — cli_verb_count (79→84), test_file_count (122→123)
+- fix(docs): wire stale action counts in README, WHITEPAPER, CLAUDE.md (78/64/79/112 actions → 116; CLI verbs 71/79 → 84)
+- feat(readme): add "What Forge Does For You" section — token savings ledger, emergent scan, BYOLLM, enforcement-at-AI-velocity
+- chore(swarm): verified recon_swarm + emergent_swarm fully wired in MCP explore tool; CLI verbs registered; real implementation confirmed
+
+## v0.84.1 — 2026-05-08
+
+- fix(mcp): add missing `items` to `repos` and `queries` array schemas in `explore` tool — resolves VS Code MCP validator rejection
+
+
+## v0.84.0 — 2026-05-08
+
+- feat(forge): v0.84 — full polyglot graft + dynamic capabilities SoT
+
+
 ## v0.83.0 — 2026-05-08
 
-- feat(forge): v0.83 — JS/TS graft support; arm Atomadic with capability transfer
+- feat(forge): v0.83 — JS/TS graft support; arm Atomadic with capability transfer
 
 
 ## v0.82.0 — 2026-05-08
 
-- feat(forge): v0.82 — graft sibling-ref auto-import + 2 grafted helpers
-- docs: README + CHANGELOG + docs/ updated for v0.76–v0.81
+- feat(forge): v0.82 — graft sibling-ref auto-import + 2 grafted helpers
+- docs: README + CHANGELOG + docs/ updated for v0.76–v0.81
 
 
 ## v0.81.0 — 2026-05-08
@@ -78,72 +109,72 @@
 ### Changed
 
 - 7f63594 feat(metrics): add repo public surface sync (GP Lane C W6)
-- d8ac951 docs: sync README + metrics for v0.75.0 — 300-seed proof banner
+- d8ac951 docs: sync README + metrics for v0.75.0 — 300-seed proof banner
 
 
 ## v0.75.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯🎯🎯🎯 HOLY GRAIL: 300 SEEDS ABSORBED (v0.75.0)
+- docs: HARVEST_LEDGER — 🎯🎯🎯🎯 HOLY GRAIL: 300 SEEDS ABSORBED (v0.75.0)
 
 
 ## v0.74.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 275 seeds (v0.74.0)
+- docs: HARVEST_LEDGER — 275 seeds (v0.74.0)
 
 
 ## v0.73.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯🎯🎯 QUARTER-THOUSAND: 250 SEEDS (v0.73.0)
+- docs: HARVEST_LEDGER — 🎯🎯🎯 QUARTER-THOUSAND: 250 SEEDS (v0.73.0)
 
 
 ## v0.72.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — v0.72.0 milestone, 225 seeds
+- docs: HARVEST_LEDGER — v0.72.0 milestone, 225 seeds
 
 
 ## v0.71.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯🎯 BICENTENNIAL: 200 SEEDS ABSORBED (v0.71.0)
+- docs: HARVEST_LEDGER — 🎯🎯 BICENTENNIAL: 200 SEEDS ABSORBED (v0.71.0)
 
 
 ## v0.70.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯 175-seed milestone (v0.70.0)
+- docs: HARVEST_LEDGER — 🎯 175-seed milestone (v0.70.0)
 
 
 ## v0.69.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯 150-seed milestone (v0.69.0)
+- docs: HARVEST_LEDGER — 🎯 150-seed milestone (v0.69.0)
 
 
 ## v0.68.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 125-seed milestone (v0.68.0)
+- docs: HARVEST_LEDGER — 125-seed milestone (v0.68.0)
 
 
 ## v0.67.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER — 🎯 century mark, 101 seeds absorbed
+- docs: HARVEST_LEDGER — 🎯 century mark, 101 seeds absorbed
 
 
 ## v0.66.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER refresh — 97 absorbed, full domain coverage
+- docs: HARVEST_LEDGER refresh — 97 absorbed, full domain coverage
 
 
 ## v0.65.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER refresh — 73 seeds, 31 at 100/100, 0 below 60
+- docs: HARVEST_LEDGER refresh — 73 seeds, 31 at 100/100, 0 below 60
 
 
 ## v0.64.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER refresh — 50+ seeds, fastapi/cookiecutter/copier landed
+- docs: HARVEST_LEDGER refresh — 50+ seeds, fastapi/cookiecutter/copier landed
 
 
 ## v0.63.0 — 2026-05-08
 
-- docs: HARVEST_LEDGER.md — proof the prompt-to-shippable spec is met
+- docs: HARVEST_LEDGER.md — proof the prompt-to-shippable spec is met
 
 
 ## v0.62.0 — 2026-05-08
@@ -153,7 +184,7 @@
 
 ## v0.61.0 — 2026-05-08
 
-- fix(forge): close audit findings — doc-update schema drift + GAP-009/010 + CHANGELOG gate
+- fix(forge): close audit findings — doc-update schema drift + GAP-009/010 + CHANGELOG gate
 
 
 ## v0.60.0 — 2026-05-08 — Forge spec met: 100/100 absorbed packages + auto-repair
@@ -270,7 +301,7 @@
 
 ### Changed
 
-- 582dc31 fix(ci): repair v0.57.0 CI failures — metrics, changelog, surface
+- 582dc31 fix(ci): repair v0.57.0 CI failures — metrics, changelog, surface
 
 
 ## v0.57.0 — 2026-05-08
@@ -1230,7 +1261,7 @@ classifier is now a true polyglot architecture engine.
   cp1252 (was breaking ``make release`` step 2).
 - ``scripts/version-sync-check.sh`` — paths are now passed to embedded
   Python via ``sys.argv[1]`` instead of interpolated into source code,
-  so ``!!AtomadicStandard``-style paths can't be misread as Python
+  so ``0_AtomadicStandard``-style paths can't be misread as Python
   escape sequences (was breaking the cross-repo sync check on Windows).
 
 ### Score
@@ -1325,7 +1356,7 @@ clean paths.
   and require an explicit ``scan`` subcommand. Both pipes now run
   end-to-end and emit producer-payload-keys summaries.
 
-### Stress-test evidence captured at C:\!!AtomadicStandard\Demo-Rehearsal\
+### Stress-test evidence captured at C:\0_AtomadicStandard\Demo-Rehearsal\
 
 - ``forge welcome`` on three Python repos: 2635 / 1854 / 7592 symbols.
 - ``forge emergent_swarm`` across forge+langgraph+crewai: 16 cross-repo

{atomadic_forge-0.83.0 → atomadic_forge-0.87.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: atomadic-forge
-Version: 0.83.0
-Summary: Atomadic Forge — absorb, enforce, emerge. Polyglot (Python · JavaScript/TypeScript · Rust · Go · Swift · Kotlin) architecture guardian for AI-generated code.
+Version: 0.87.0
+Summary: Atomadic Forge ? absorb, enforce, emerge. Polyglot (Python ? JavaScript/TypeScript ? Rust ? Go ? Swift ? Kotlin) architecture guardian for AI-generated code.
 Author: Atomadic
 License-Expression: BUSL-1.1
 Keywords: atomadic,ass-ade,monadic,architecture,forge,code-quality,ai,agents,javascript,typescript,polyglot
@@ -89,6 +89,24 @@ AI agents write code **10x faster** than humans can review it. Pattern violation
 
 ---
 
+## What Forge Does For You
+
+Forge is an architecture enforcement engine, not a linter. It imposes deterministic structure on AI-generated code — and it ships proof that the structure holds.
+
+- **Architectural drift stops here.** `forge auto` runs recon, detects every upward-import violation, re-tiers mislaid files, and emits a signed conformance receipt — all in one command. Validated against 300+ OSS repositories: Wire-PASS holds across the entire corpus.
+
+- **Context overhead you can actually measure.** `forge context_pack` pre-bundles repo context into a single optimized payload for AI agents, cutting redundant re-reads. Every Forge verb logs its savings to a file-backed JSONL ledger at `<project>/.atomadic-forge/lifetime_savings.jsonl` — run `forge token-savings` to see the auditable running total, line by line.
+
+- **Composition opportunities no human would find.** `forge emergent scan` performs cross-repo AST analysis to surface capability combinations that exist in your codebase but have never been wired together. Discovery is deterministic — no LLM guessing, no hallucination false positives.
+
+- **One command: spaghetti to certified.** `forge auto` takes any Python, JS, TS, Rust, Go, Swift, or Kotlin repo from chaos to 5-tier-organized, 0 wire violations, 100/100 certify score. The before/after is not a demo — receipts are in [`HARVEST_LEDGER.md`](HARVEST_LEDGER.md).
+
+- **Bring your own LLM — any of them.** OpenAI, Anthropic, Gemini, OpenRouter, Ollama, DeepSeek, Groq, Mistral, and 10+ more ship as built-in presets. Add any OpenAI-compatible endpoint with `forge config provider add`. No vendor lock-in, no rewiring required.
+
+- **Enforcement that scales with AI output velocity.** Forge's tier-law scanner is pure AST — no LLM in the loop, no nondeterminism. Promote violations from advisory to `--fail-on-violations` and it becomes a hard CI gate. Architecture holds at the speed AI generates code.
+
+---
+
 ## Quick Start
 
 ```bash
@@ -192,11 +210,11 @@ Forge enforces a **5-tier monadic law** — every source file belongs to exactly
 | `forge plan` | Bounded execution plan with risk + rollback |
 | `forge chat` | Terminal copilot over forge docs |
 
-**79 CLI verbs total.** Full reference: [docs/02-commands.md](docs/02-commands.md)
+Live CLI verb count: see [forge_metrics.json](forge_metrics.json). Full reference: [docs/02-commands.md](docs/02-commands.md)
 
 ---
 
-## MCP Server — 10 Tools, 116 Actions
+## MCP Server
 
 Forge ships a Model Context Protocol server. Add it to Cursor, Claude Code, Aider, Devin, or any MCP-compatible agent:
 
@@ -211,20 +229,21 @@ Forge ships a Model Context Protocol server. Add it to Cursor, Claude Code, Aide
 }
 ```
 
-**10 tools, 116 actions locally.** All legacy tool names redirect automatically — existing callers keep working. Deployed Worker at [forge.atomadic.tech/mcp](https://forge.atomadic.tech/mcp). Listed in the [MCP Official Registry](https://registry.modelcontextprotocol.io/).
-
-| Tool | Actions | What it does |
-|------|---------|-------------|
-| `welcome` | — | RECOMMENDED FIRST CALL: onboarding scan, score, narrative, capability tour |
-| `explore` | 17 | Analyze, understand, discover — recon/explain/call_graph/call_graph_summary/smell/lineage/harvest/synergy/scan/swarm/churn/surface_gaps/capability_scout/dead_code/maintainability/concept_bridge |
-| `audit` | 23 | Quality gates, change safety, trust |
-| `plan` | 10 | Orient + dev utilities — context/compose/policy/recipes/generate/apply/locate/commit/scaffold/steersman |
-| `transmute` | 3 | **FLAGSHIP** spaghetti→certified pipeline — auto/cherry/finalize |
-| `loop` | 4 | LLM iteration loops — iterate/resume/evolve/evolve_step |
-| `hive` | 13 | Multi-agent coordination — lifecycle, consensus, handoffs, enhancements |
-| `wisdom` | 5 | Institutional memory DB — record/query/list/recall/promote |
-| `nexus` | 7 | AAAA-Nexus auth primitives (uses `op=` not `action=`) — identity/federation/authorize/trust/contract/lineage/ratchet |
-| `create` | — | Intent + seed repos → shippable pip package (Phase 1 pipeline: emergent_swarm + materialize) |
+Current live surface: see [surface.json](surface.json) and [forge_metrics.json](forge_metrics.json). All legacy tool names redirect automatically — existing callers keep working. Deployed Worker at [forge.atomadic.tech/mcp](https://forge.atomadic.tech/mcp). Listed in the [MCP Official Registry](https://registry.modelcontextprotocol.io/).
+
+| Tool | Dispatch | What it does |
+|------|----------|-------------|
+| `welcome` | none | RECOMMENDED FIRST CALL: onboarding scan, score, narrative, capability tour |
+| `explore` | `action=` | Analyze, understand, discover — recon/explain/call_graph/call_graph_summary/smell/lineage/harvest/synergy/scan/swarm/churn/surface_gaps/capability_scout/dead_code/maintainability/concept_bridge |
+| `audit` | `action=` | Quality gates, change safety, trust |
+| `plan` | `action=` | Orient + dev utilities — context/compose/policy/recipes/generate/apply/locate/commit/scaffold/steersman |
+| `transmute` | `action=` | **FLAGSHIP** spaghetti→certified pipeline — auto/cherry/finalize |
+| `loop` | `action=` | LLM iteration loops — iterate/resume/evolve/evolve_step |
+| `hive` | `action=` | Multi-agent coordination — lifecycle, consensus, handoffs, enhancements |
+| `wisdom` | `action=` | Institutional memory DB — record/query/list/recall/promote |
+| `nexus` | `op=` | AAAA-Nexus auth primitives — identity/federation/authorize/trust/contract/lineage/ratchet |
+| `inspect` | `action=` | Low-level AST inspection — body/symbol/class_signals/ast_scope_blueprint/mint_ast_locks/add_node/classify_repo_layout/fingerprint_repo/synthesis_plan/swarm_repo_specs |
+| `create` | none | Intent + seed repos → shippable pip package (Phase 1 pipeline: emergent_swarm + materialize) |
 
 > **Note:** `nexus` uses `op=` as its dispatch key instead of `action=` to avoid collision with the `authorize_action` operation's own `action` parameter that passes through to the Nexus API.
 
@@ -234,21 +253,14 @@ Full tool reference: [docs/MCP_TOOLS.md](docs/MCP_TOOLS.md) | First call: `welco
 
 ## By the Numbers
 
-> Every number below is a live read from [`forge_metrics.json`](forge_metrics.json) — the canonical metrics artifact regenerated on every push by `forge metrics --apply`. CI fails any PR that lets the file drift.
+> Live repo metrics come from [forge_metrics.json](forge_metrics.json). Live MCP surface shape and legacy redirects come from [surface.json](surface.json). Historical release counts stay in [CHANGELOG.md](CHANGELOG.md) and [receipts/](receipts/).
 
 | Metric | Value |
 |--------|-------|
 
-| MCP tools | **10 tools · 116 actions** |
-| CLI verbs | **79** |
-| Source files | **246** (~49K LOC) |
-| Tests | **1625 passing** across **121 test files** |
-| Lean4 formal proofs | **872 theorems** (mathematically verified core logic) |
-| Monadic tiers | **5** (a0 → a4, strict upward-only composition) |
-| Languages | Python, JS/TS, Swift, Kotlin, Go, Rust |
-| Infrastructure cost | **$5/mo** |
-| Self-score | **100/100** certify · 0 wire violations |
-| License | BSL 1.1 (Apache 2.0 after 2030-04-27) |
+| Current version, CLI, source, test, and proof counts | [`forge_metrics.json`](forge_metrics.json) |
+| Current MCP tools, actions, recipes, and legacy redirects | [`surface.json`](surface.json) |
+| Historical release snapshots | [`CHANGELOG.md`](CHANGELOG.md), [`receipts/`](receipts/) |
 
 Forge eats its own cooking. Every CI run scores itself.
 
@@ -265,7 +277,7 @@ Forge eats its own cooking. Every CI run scores itself.
 |------|-------|-------------|
 | **Free** | $0 | OSS, 25 calls/day, read-only tools |
 | **One-Time Refactor** | ~~$149~~ **$99** | We run `auto` on your repo -> PR targeting 100/100. One revision included. |
-| **Starter** | $79/mo | All 10 tools (112 actions), 5k calls/mo, signed receipts, custom recipes |
+| **Starter** | $79/mo | Full current Forge surface, 5k calls/mo, signed receipts, custom recipes |
 | **Pro** | $249/mo | + notarization, compliance attestations, cross-repo telemetry |
 | **Founding Member** | $999/yr | Everything. Forever. First 25 only. |
 | **Enterprise** | Custom | BSL commercial license, SSO, self-host, SLA |

{atomadic_forge-0.83.0 → atomadic_forge-0.87.0}/README.md

@@ -57,6 +57,24 @@ AI agents write code **10x faster** than humans can review it. Pattern violation
 
 ---
 
+## What Forge Does For You
+
+Forge is an architecture enforcement engine, not a linter. It imposes deterministic structure on AI-generated code — and it ships proof that the structure holds.
+
+- **Architectural drift stops here.** `forge auto` runs recon, detects every upward-import violation, re-tiers mislaid files, and emits a signed conformance receipt — all in one command. Validated against 300+ OSS repositories: Wire-PASS holds across the entire corpus.
+
+- **Context overhead you can actually measure.** `forge context_pack` pre-bundles repo context into a single optimized payload for AI agents, cutting redundant re-reads. Every Forge verb logs its savings to a file-backed JSONL ledger at `<project>/.atomadic-forge/lifetime_savings.jsonl` — run `forge token-savings` to see the auditable running total, line by line.
+
+- **Composition opportunities no human would find.** `forge emergent scan` performs cross-repo AST analysis to surface capability combinations that exist in your codebase but have never been wired together. Discovery is deterministic — no LLM guessing, no hallucination false positives.
+
+- **One command: spaghetti to certified.** `forge auto` takes any Python, JS, TS, Rust, Go, Swift, or Kotlin repo from chaos to 5-tier-organized, 0 wire violations, 100/100 certify score. The before/after is not a demo — receipts are in [`HARVEST_LEDGER.md`](HARVEST_LEDGER.md).
+
+- **Bring your own LLM — any of them.** OpenAI, Anthropic, Gemini, OpenRouter, Ollama, DeepSeek, Groq, Mistral, and 10+ more ship as built-in presets. Add any OpenAI-compatible endpoint with `forge config provider add`. No vendor lock-in, no rewiring required.
+
+- **Enforcement that scales with AI output velocity.** Forge's tier-law scanner is pure AST — no LLM in the loop, no nondeterminism. Promote violations from advisory to `--fail-on-violations` and it becomes a hard CI gate. Architecture holds at the speed AI generates code.
+
+---
+
 ## Quick Start
 
 ```bash
@@ -160,11 +178,11 @@ Forge enforces a **5-tier monadic law** — every source file belongs to exactly
 | `forge plan` | Bounded execution plan with risk + rollback |
 | `forge chat` | Terminal copilot over forge docs |
 
-**79 CLI verbs total.** Full reference: [docs/02-commands.md](docs/02-commands.md)
+Live CLI verb count: see [forge_metrics.json](forge_metrics.json). Full reference: [docs/02-commands.md](docs/02-commands.md)
 
 ---
 
-## MCP Server — 10 Tools, 116 Actions
+## MCP Server
 
 Forge ships a Model Context Protocol server. Add it to Cursor, Claude Code, Aider, Devin, or any MCP-compatible agent:
 
@@ -179,20 +197,21 @@ Forge ships a Model Context Protocol server. Add it to Cursor, Claude Code, Aide
 }
 ```
 
-**10 tools, 116 actions locally.** All legacy tool names redirect automatically — existing callers keep working. Deployed Worker at [forge.atomadic.tech/mcp](https://forge.atomadic.tech/mcp). Listed in the [MCP Official Registry](https://registry.modelcontextprotocol.io/).
-
-| Tool | Actions | What it does |
-|------|---------|-------------|
-| `welcome` | — | RECOMMENDED FIRST CALL: onboarding scan, score, narrative, capability tour |
-| `explore` | 17 | Analyze, understand, discover — recon/explain/call_graph/call_graph_summary/smell/lineage/harvest/synergy/scan/swarm/churn/surface_gaps/capability_scout/dead_code/maintainability/concept_bridge |
-| `audit` | 23 | Quality gates, change safety, trust |
-| `plan` | 10 | Orient + dev utilities — context/compose/policy/recipes/generate/apply/locate/commit/scaffold/steersman |
-| `transmute` | 3 | **FLAGSHIP** spaghetti→certified pipeline — auto/cherry/finalize |
-| `loop` | 4 | LLM iteration loops — iterate/resume/evolve/evolve_step |
-| `hive` | 13 | Multi-agent coordination — lifecycle, consensus, handoffs, enhancements |
-| `wisdom` | 5 | Institutional memory DB — record/query/list/recall/promote |
-| `nexus` | 7 | AAAA-Nexus auth primitives (uses `op=` not `action=`) — identity/federation/authorize/trust/contract/lineage/ratchet |
-| `create` | — | Intent + seed repos → shippable pip package (Phase 1 pipeline: emergent_swarm + materialize) |
+Current live surface: see [surface.json](surface.json) and [forge_metrics.json](forge_metrics.json). All legacy tool names redirect automatically — existing callers keep working. Deployed Worker at [forge.atomadic.tech/mcp](https://forge.atomadic.tech/mcp). Listed in the [MCP Official Registry](https://registry.modelcontextprotocol.io/).
+
+| Tool | Dispatch | What it does |
+|------|----------|-------------|
+| `welcome` | none | RECOMMENDED FIRST CALL: onboarding scan, score, narrative, capability tour |
+| `explore` | `action=` | Analyze, understand, discover — recon/explain/call_graph/call_graph_summary/smell/lineage/harvest/synergy/scan/swarm/churn/surface_gaps/capability_scout/dead_code/maintainability/concept_bridge |
+| `audit` | `action=` | Quality gates, change safety, trust |
+| `plan` | `action=` | Orient + dev utilities — context/compose/policy/recipes/generate/apply/locate/commit/scaffold/steersman |
+| `transmute` | `action=` | **FLAGSHIP** spaghetti→certified pipeline — auto/cherry/finalize |
+| `loop` | `action=` | LLM iteration loops — iterate/resume/evolve/evolve_step |
+| `hive` | `action=` | Multi-agent coordination — lifecycle, consensus, handoffs, enhancements |
+| `wisdom` | `action=` | Institutional memory DB — record/query/list/recall/promote |
+| `nexus` | `op=` | AAAA-Nexus auth primitives — identity/federation/authorize/trust/contract/lineage/ratchet |
+| `inspect` | `action=` | Low-level AST inspection — body/symbol/class_signals/ast_scope_blueprint/mint_ast_locks/add_node/classify_repo_layout/fingerprint_repo/synthesis_plan/swarm_repo_specs |
+| `create` | none | Intent + seed repos → shippable pip package (Phase 1 pipeline: emergent_swarm + materialize) |
 
 > **Note:** `nexus` uses `op=` as its dispatch key instead of `action=` to avoid collision with the `authorize_action` operation's own `action` parameter that passes through to the Nexus API.
 
@@ -202,21 +221,14 @@ Full tool reference: [docs/MCP_TOOLS.md](docs/MCP_TOOLS.md) | First call: `welco
 
 ## By the Numbers
 
-> Every number below is a live read from [`forge_metrics.json`](forge_metrics.json) — the canonical metrics artifact regenerated on every push by `forge metrics --apply`. CI fails any PR that lets the file drift.
+> Live repo metrics come from [forge_metrics.json](forge_metrics.json). Live MCP surface shape and legacy redirects come from [surface.json](surface.json). Historical release counts stay in [CHANGELOG.md](CHANGELOG.md) and [receipts/](receipts/).
 
 | Metric | Value |
 |--------|-------|
 
-| MCP tools | **10 tools · 116 actions** |
-| CLI verbs | **79** |
-| Source files | **246** (~49K LOC) |
-| Tests | **1625 passing** across **121 test files** |
-| Lean4 formal proofs | **872 theorems** (mathematically verified core logic) |
-| Monadic tiers | **5** (a0 → a4, strict upward-only composition) |
-| Languages | Python, JS/TS, Swift, Kotlin, Go, Rust |
-| Infrastructure cost | **$5/mo** |
-| Self-score | **100/100** certify · 0 wire violations |
-| License | BSL 1.1 (Apache 2.0 after 2030-04-27) |
+| Current version, CLI, source, test, and proof counts | [`forge_metrics.json`](forge_metrics.json) |
+| Current MCP tools, actions, recipes, and legacy redirects | [`surface.json`](surface.json) |
+| Historical release snapshots | [`CHANGELOG.md`](CHANGELOG.md), [`receipts/`](receipts/) |
 
 Forge eats its own cooking. Every CI run scores itself.
 
@@ -233,7 +245,7 @@ Forge eats its own cooking. Every CI run scores itself.
 |------|-------|-------------|
 | **Free** | $0 | OSS, 25 calls/day, read-only tools |
 | **One-Time Refactor** | ~~$149~~ **$99** | We run `auto` on your repo -> PR targeting 100/100. One revision included. |
-| **Starter** | $79/mo | All 10 tools (112 actions), 5k calls/mo, signed receipts, custom recipes |
+| **Starter** | $79/mo | Full current Forge surface, 5k calls/mo, signed receipts, custom recipes |
 | **Pro** | $249/mo | + notarization, compliance attestations, cross-repo telemetry |
 | **Founding Member** | $999/yr | Everything. Forever. First 25 only. |
 | **Enterprise** | Custom | BSL commercial license, SSO, self-host, SLA |

{atomadic_forge-0.83.0 → atomadic_forge-0.87.0}/docs/05-faq.md

@@ -518,27 +518,13 @@ config:
 }
 ```
 
-Once registered (Forge ≥ v0.14.0a10), the agent gets **61 tools** + **5
-resources**. The 61 tools group into *inventory / utility* (`recon` /
-`wire` / `certify` / `verify` / `enforce` / `lineage` /
-`emergent_scan` / `emergent_swarm` / `synergy_scan` / `doctor` /
-`sidecar_validate` / `manifest_diff` / `call_graph` / `smell_scan` /
-`wisdom_record` + `wisdom_query` + `wisdom_list` + `wisdom_recall` /
-`surface`), *action loop* (`plan` / `plan_apply`), *Copilot's
-Copilot* (`context_pack` / `preflight_change` / `score_patch` /
-`select_tests` / `rollback_plan` / `explain_repo` / `compose_tools` /
-`load_policy` / `recipes` / `forge_locate` / `commit_compose` /
-`cna_check` / `trust_gate_response`), *transmuter / cross-repo*
-(`auto` / `cherry` / `finalize` / `iterate_start` /
-`iterate_continue` / `evolve_start` / `evolve_step` / `harvest` /
-`recon_swarm` / `tool_factory` / `guard_install`), *hive
-coordination* (`hive_register` / `hive_observe` / `hive_propose` /
-`hive_vote` / `hive_result` / `hive_list` / `hive_deactivate` /
-`hive_needs_vote` / `hive_recap` since v0.14.0a9–v0.14.0a10), and
-*AAAA-Nexus primitives* (`nexus_identity_verify` / `nexus_federation_mint` /
-`nexus_authorize_action` / `nexus_sys_trust_gate` / `nexus_contract_verify` /
-`nexus_lineage_record` / `nexus_ratchet_register` since v0.14.0a7).
-See the full **[Agents Guide](AGENTS_GUIDE.md)** for
+Once registered, the agent gets the current live umbrella surface recorded in
+[`../forge_metrics.json`](../forge_metrics.json) and [`surface.json`](../surface.json),
+with legacy endpoint names resolving through `surface.json -> legacy_endpoint_map`.
+For day-to-day work, agents mostly live in `welcome`, `explore`,
+`audit`, and `plan`; the deeper lanes are `inspect`, `transmute`,
+`loop`, `hive`, `wisdom`, `nexus`, and `create`. See the full
+**[Agents Guide](AGENTS_GUIDE.md)** for
 integration patterns, the agent-friendly `context_pack` and
 `explain_repo` tools, the Forge Receipt schema agents should
 consume, F-code routing for mechanical fixes, the proposal-engine