Skip to content

"CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: https://clianything.cc/

Python

fix: correct CodeGraph tool names and remove isolation promise

Locodocowwf0560fc73616
diff --git a/README.md b/README.md
index df9c1765d..1e255560f 100644
--- a/README.md
+++ b/README.md
@@ -651,7 +651,7 @@ Use CLI-Anything to validate ./libreoffice
The Reasonix skill adapts the same methodology used by the Claude Code plugin and
Codex/Hermes skills, binding Reasonix's `bash`, `write_file`, `edit_file`,
-`multi_edit`, `grep`, `glob`, and optional `mcp__codegraph__*` tools to the
+`multi_edit`, `grep`, `glob`, and optional `mcp__codegraph__search` / `mcp__codegraph__context` tools to the
7-phase harness workflow while keeping the generated Python harness format unchanged.
</details>
diff --git a/reasonix-skill/README.md b/reasonix-skill/README.md
index 4b42df383..6b085a3ba 100644
--- a/reasonix-skill/README.md
+++ b/reasonix-skill/README.md
@@ -40,7 +40,7 @@ run_skill({ name: "cli-anything", arguments: "/path/to/software" })
### What It Does
-The skill spawns an isolated Reasonix sub-agent that follows the CLI-Anything 7-phase methodology:
+The skill follows the CLI-Anything 7-phase methodology:
1. **Codebase Analysis** — Surveys the target software's architecture, data model, and API surface
2. **CLI Architecture Design** — Designs command groups, state model, and output formats
@@ -72,7 +72,7 @@ The skill spawns an isolated Reasonix sub-agent that follows the CLI-Anything 7-
| Code editing | `edit_file` / `multi_edit` |
| Source reading | `read_file` |
| Codebase search | `grep` / `glob` / `ls` |
-| Architecture analysis | `mcp__codegraph__*` when CodeGraph is enabled |
+| Architecture analysis | `mcp__codegraph__search` / `mcp__codegraph__context` when CodeGraph is enabled |
| Documentation fetch | `web_fetch` |
## File Structure
@@ -95,7 +95,7 @@ reasonix-skill/
| Skill format | `.claude-plugin/` | `SKILL.md` + YAML | `SKILL.md` + YAML | `SKILL.md` + YAML |
| Shell | `bash` | `terminal` | `terminal` | `bash` |
| File ops | native | `execute_code` | `execute_code` / `write_file` | `write_file` / `edit_file` / `multi_edit` |
-| Code analysis | file reads | file reads | file reads | Optional `mcp__codegraph__*` symbol graph |
+| Code analysis | file reads | file reads | file reads | Optional `mcp__codegraph__search` / `mcp__codegraph__context` symbol graph |
| Install method | `/plugin install` | `install.sh` → `$CODEX_HOME/skills/` | `install.sh` → `$HERMES_HOME/skills/` | `install.sh` → `~/.reasonix/skills/` |
## Contributing
diff --git a/reasonix-skill/SKILL.md b/reasonix-skill/SKILL.md
index 961b99cd6..c91d479b0 100644
--- a/reasonix-skill/SKILL.md
+++ b/reasonix-skill/SKILL.md
@@ -38,12 +38,12 @@ Reasonix agents build harnesses by combining these built-in tools:
| `grep` | Search for patterns across the target software codebase (APIs, CLI tools, data models) |
| `glob` | Find files matching patterns in the source tree (*.py, *.xml, *.json, etc.) |
| `ls` | List directory contents to understand project structure |
-| `mcp__codegraph__*` | Optional code graph analysis when CodeGraph is enabled |
+| `mcp__codegraph__search` / `mcp__codegraph__context` | Optional code graph analysis when CodeGraph is enabled (Reasonix strips the `codegraph_` raw prefix, so the model-visible names omit it) |
| `web_fetch` | Fetch documentation, API references, or remote files from the web |
### Recommended Workflow with Reasonix Tools
-**Phase 1 (Analysis)** — Use `ls` + `glob` to survey the source tree, `grep` to find API surfaces and CLI entry points, and `read_file` to inspect key files. When CodeGraph is enabled and its tools are available, use `mcp__codegraph__codegraph_search` and `mcp__codegraph__codegraph_context` for deeper symbol and architecture analysis.
+**Phase 1 (Analysis)** — Use `ls` + `glob` to survey the source tree, `grep` to find API surfaces and CLI entry points, and `read_file` to inspect key files. When CodeGraph is enabled and its tools are available, use `mcp__codegraph__search` and `mcp__codegraph__context` for deeper symbol and architecture analysis.
**Phase 2-3 (Design & Implementation)** — Use `write_file` to create new harness files, `edit_file` / `multi_edit` to refine generated code, and `bash` to run `pip install -e .` for installation.