docs(mcp): document postprocess operations surface

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

Author: tae2089

README.md

@@ -1,13 +1,13 @@
 # code-context-graph
 
-Local code analysis tool that parses codebases via Tree-sitter into a knowledge graph. Supports 12 languages, 31 MCP tools, and custom annotation search.
+Local code analysis tool that parses codebases via Tree-sitter into a knowledge graph. Supports 12 languages, 33 MCP tools, and custom annotation search.
 
 Inspired by [code-review-graph](https://github.com/tirth8205/code-review-graph) — a Python-based code analysis tool. This project reimplements and extends the concept in Go with multi-DB support, custom annotation system, and MCP integration for AI-powered code understanding.
 
 ## Features
 
 - **12 languages**: Go, Python, TypeScript, Java, Ruby, JavaScript, C, C++, Rust, Kotlin, PHP, Lua/Luau
-- **31 MCP tools**: parse, search, impact analysis, flow tracing, dead code detection, namespace file management, and more
+- **33 MCP tools**: parse, search, impact analysis, flow tracing, dead code detection, postprocess operations, namespace file management, and more
 - **Custom annotations**: `@intent`, `@domainRule`, `@sideEffect`, `@mutates`, `@index` — search code by business context ([details](guide/annotations.md))
 - **Webhook sync**: GitHub / Gitea push events → auto clone + build with per-repo branch filtering and `.ccg.yaml` `include_paths` auto-loading ([details](guide/webhook.md))
 - **Eval**: Golden corpus-based parser accuracy (P/R/F1) and search quality (P@K, MRR, nDCG) evaluation ([details](guide/cli-reference.md#eval))
@@ -176,7 +176,7 @@ For remote HTTP mode:
 }
 ```
 
-Claude Code automatically connects and gets access to 31 MCP tools. See [MCP Tools Reference](guide/mcp-tools.md) for the full list.
+Claude Code automatically connects and gets access to 33 MCP tools. See [MCP Tools Reference](guide/mcp-tools.md) for the full list.
 
 ## Architecture
 
@@ -187,7 +187,7 @@ Source Code → Tree-sitter Parser → Nodes + Edges + Annotations
                                         ↓
                                    FTS Search
                                         ↓
-                              MCP Server (31 tools)
+                              MCP Server (33 tools)
                                     ↓         ↓
                               stdio       Streamable HTTP
                                 ↓              ↓
@@ -204,7 +204,7 @@ See [Architecture Details](guide/architecture.md) for component breakdown and DB
 | Guide | Description |
 |-------|-------------|
 | [CLI Reference](guide/cli-reference.md) | All commands, flags, and config file (`.ccg.yaml`) |
-| [MCP Tools](guide/mcp-tools.md) | 31 MCP tools, Skills, AI-Driven Annotation |
+| [MCP Tools](guide/mcp-tools.md) | 33 MCP tools, Skills, AI-Driven Annotation |
 | [Annotations](guide/annotations.md) | Annotation system — tags, examples, search |
 | [Webhook](guide/webhook.md) | Webhook sync, branch filtering, HMAC, graceful shutdown |
 | [Docker](guide/docker.md) | Docker build, MCP server, PostgreSQL deployment |

guide/README.md

@@ -5,7 +5,7 @@ Documentation index for code-context-graph.
 | Document | Description |
 |----------|-------------|
 | [CLI Reference](cli-reference.md) | Full CLI commands, options, and configuration file (`.ccg.yaml`) |
-| [MCP Tools](mcp-tools.md) | 31 MCP tools, Claude Code Skills, AI-driven annotation |
+| [MCP Tools](mcp-tools.md) | 33 MCP tools, Claude Code Skills, AI-driven annotation |
 | [Annotations](annotations.md) | Custom annotation system — tags, examples, search |
 | [Webhook](webhook.md) | GitHub / Gitea webhook sync, branch filtering, graceful shutdown |
 | [Docker](docker.md) | Docker image build, MCP server setup, PostgreSQL integration |

guide/architecture.md

@@ -9,7 +9,7 @@ Source Code → Tree-sitter Parser → Nodes + Edges + Annotations
                                         ↓
                                    FTS Search
                                         ↓
-                              MCP Server (31 tools)
+                              MCP Server (33 tools)
                                     ↓         ↓
                               stdio       Streamable HTTP
                                 ↓              ↓
@@ -77,7 +77,7 @@ Golden corpus-based parser accuracy and search quality evaluation framework.
 
 ### MCP Server (`internal/mcp/`)
 
-Exposes 31 tools via MCP protocol. Supports two transport modes: stdio and Streamable HTTP.
+Exposes 33 tools via MCP protocol. Supports two transport modes: stdio and Streamable HTTP.
 
 ### Reliability
 

guide/development.md

@@ -104,7 +104,7 @@ internal/
   ctxns/              — Context namespace
   docs/               — Documentation generation
   eval/               — Parser/search quality evaluation (golden corpus, P/R/F1, P@K, MRR, nDCG)
-  mcp/                — MCP server (29 tools)
+  mcp/                — MCP server (33 tools)
   model/              — DB models
   parse/treesitter/   — Tree-sitter parser (12 languages, including Lua/Luau)
   pathutil/           — Path utilities

guide/mcp-tools.md

@@ -1,6 +1,6 @@
 # MCP Tools
 
-code-context-graph provides 31 MCP tools. Automatically connects from Claude Code after configuring `.mcp.json`.
+code-context-graph provides 33 MCP tools. Automatically connects from Claude Code after configuring `.mcp.json`.
 
 ## Setup
 
@@ -30,7 +30,7 @@ code-context-graph provides 31 MCP tools. Automatically connects from Claude Cod
 }
 ```
 
-## Tools (31)
+## Tools (33)
 
 ### Core
 
@@ -39,6 +39,8 @@ code-context-graph provides 31 MCP tools. Automatically connects from Claude Cod
 | `parse_project` | Parse source files |
 | `build_or_update_graph` | Full/incremental build with postprocessing |
 | `run_postprocess` | Rebuild stored flows, communities, and/or full-text search derived state |
+| `get_postprocess_policy` | Inspect automatic postprocess policy state and recent failures |
+| `reset_postprocess_policy` | Record a reset marker to clear fail-closed streak for one tool |
 | `get_node` | Get node by qualified name |
 | `search` | Full-text search |
 | `query_graph` | Predefined graph queries (callers, callees, imports, etc.) |
@@ -54,6 +56,10 @@ See [Postprocess Failure Policy](postprocess-failure-policy.md) for the detailed
 status tables, failure causes, skip behavior, and policy escalation rules for
 `build_or_update_graph` and `run_postprocess`.
 
+CCG does not expose a Prometheus `/metrics` endpoint yet. For postprocess
+operations, use `get_postprocess_policy` and the HTTP `/status` summary as the
+current machine-readable operational surfaces.
+
 ### Analysis
 
 | Tool | Description |

guide/postprocess-failure-policy.md

@@ -37,6 +37,18 @@ Policy state is persisted as:
 - current effective state in `ccg_postprocess_policy_state`
 - append-only execution history in `ccg_postprocess_run_logs`
 
+For operators, CCG now exposes two lightweight control surfaces:
+
+- `get_postprocess_policy` returns the current fail-closed entries and recent failures.
+- `reset_postprocess_policy` records a reset marker run for one tool in the current namespace.
+
+The reset path does not delete execution history. It inserts an `ok` marker with
+source `reset`, which safely breaks the consecutive-failure streak used by the
+automatic resolver.
+
+Run logs are also pruned opportunistically after writes so each `(namespace,
+tool)` keeps only a bounded recent history instead of growing forever.
+
 ## Namespace isolation and shared state
 
 Automatic policy decisions and postprocess results are scoped to the current