@yawlabs/ctxlint 0.19.0 → 0.20.0

package/README.md

@@ -511,9 +511,9 @@ ctxlint is the reference implementation of four open specifications for linting
 
 | Spec                                                           | What it covers                                                                                                                                                                                                                            |
 | -------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| **[AI Context File Linting Spec](./CONTEXT_LINT_SPEC.md)**     | 39 rules for validating context files (CLAUDE.md, .cursorrules, AGENTS.md, etc.) across 16 clients. Covers file formats, frontmatter schemas, path/command validation, staleness, token budgets, redundancy, and contradictions.          |
+| **[AI Context File Linting Spec](./CONTEXT_LINT_SPEC.md)**     | 41 rules for validating context files (CLAUDE.md, .cursorrules, AGENTS.md, etc.) across 16 clients. Covers file formats, frontmatter schemas, path/command validation, staleness, token budgets, redundancy, and contradictions.          |
 | **[MCP Config Linting Spec](./MCP_CONFIG_LINT_SPEC.md)**       | 29 rules for validating MCP server configs (.mcp.json, .cursor/mcp.json, .vscode/mcp.json, etc.) across 8 clients. Covers schema validation, hardcoded secrets, env var syntax, deprecated transports, and cross-file consistency.        |
-| **[Agent Session Linting Spec](./AGENT_SESSION_LINT_SPEC.md)** | 11 rules for auditing agent session data (history, memory) across 8 agents. Covers cross-project secret consistency, config drift, stale memory, and loop detection.                                                                       |
+| **[Agent Session Linting Spec](./AGENT_SESSION_LINT_SPEC.md)** | 12 rules for auditing agent session data (history, memory) across 8 agents. Covers cross-project secret consistency, config drift, stale memory, and loop detection.                                                                       |
 | **[Agent Skill Linting Spec](./AGENT_SKILL_LINT_SPEC.md)**     | 5 rules for auditing Claude Code skill (`SKILL.md`) and agent (`.md`) definitions under `~/.claude`. Covers frontmatter presence, broken refs, trigger-phrase collisions, orphaned skills, and dead tool restrictions. (v1, experimental) |
 
 All specs include machine-readable rule catalogs for programmatic consumption:

package/agent-session-lint-rules.json

@@ -134,6 +134,16 @@
       "message": "{count} files edited on '{branch}' with no intervening commit",
       "fixable": false,
       "stability": "stable"
+    },
+    {
+      "id": "session/unresolvable-sha",
+      "category": "session",
+      "severity": "warning",
+      "description": "A memory cites a git SHA that does not resolve in the repository.",
+      "trigger": "A 7-40 character hex token outside code fences, preceded on the same line by a commit-citation cue (commit, SHA, HEAD, landed in, reverted, ...), does not resolve via git cat-file. UUID members, digest prefixes, decimal ids and version fragments are excluded before resolution; without a git repository the rule reports nothing.",
+      "message": "cited commit {sha} does not resolve in this repository",
+      "fixable": false,
+      "stability": "experimental"
     }
   ],
   "dataSources": [

package/context-lint-rules.json

@@ -170,6 +170,26 @@
       "fixable": false,
       "stability": "stable"
     },
+    {
+      "id": "commands/exit-status-masked",
+      "category": "commands",
+      "severity": "warning",
+      "description": "A verification command's exit status is discarded by a pipe into a filter, so the success claim that follows cannot fail.",
+      "trigger": "A verifier (tsc, eslint, biome, vitest, pytest, cargo test, or an npm/pnpm/yarn/bun script resolving to one) heads a pipeline whose last stage is a filter or pager (head, tail, grep, sed, awk, less, cat, tee, wc), and the pipeline is followed by && plus a success-announcing echo/printf, or by an echo of $?, with no set -o pipefail in scope.",
+      "message": "\"{cmd}\" — exit status comes from \"{filter}\", not \"{verifier}\"; the success claim cannot fail",
+      "fixable": false,
+      "stability": "experimental"
+    },
+    {
+      "id": "commands/unknown-subcommand",
+      "category": "commands",
+      "severity": "error",
+      "description": "A documented invocation of the project's own binary uses a subcommand the CLI does not implement.",
+      "trigger": "A command invokes a name declared in package.json#bin with a leading non-flag argument that is not in the CLI's known subcommand set, where that set was statically resolved as closed from the entry file.",
+      "message": "\"{cmd}\" — \"{sub}\" is not a subcommand of {bin} (known: {known})",
+      "fixable": false,
+      "stability": "experimental"
+    },
     {
       "id": "staleness/stale",
       "category": "staleness",