fix: more strict check for registry to avoid false positives

[gagdiez]

Summary

By default ironclaw gets installed in ~/.cargo/bin/ironclaw, if the user happens to try to compile anything (e.g. their own tool) then the ~/.cargo/registry folder gets created, which makes ironclaw think that it found an (empty) registry, and thus stops being able to do list or install tools from its internal registry

Change Type

• Bug fix
• New feature
• Refactor
• Documentation
• CI/Infrastructure
• Security
• Dependencies

Linked Issue

Closes #2221

Validation

• cargo fmt --all -- --check
• cargo test
• cargo clippy --all --benches --tests --examples --all-features -- -D warnings
• cargo build
• Relevant tests pass:
• cargo test --features integration if database-backed or integration behavior changed
• Manual testing: Built, created a registry folder on parent directory, and tested that ironclaw registry list still worked
• If a coding agent was used and supports it, review-pr or pr-shepherd --fix was run before requesting review

Security Impact

None

Database Impact

None

Blast Radius

Registry load

Rollback Plan

Revert is_registry to simply check for is_dir

---

Review track: A or B

[Copilot]

Pull request overview

Fixes false-positive detection of an on-disk registry/ directory (notably ~/.cargo/registry) so ironclaw registry list/install continues to work by falling back to the embedded registry when a real internal registry isn’t present.

Changes:

• Replace candidate.is_dir() checks with a stricter is_registry() predicate during registry discovery.
• Add unit tests covering is_registry() accept/reject cases.
• Update build.rs to add directory-level cargo:rerun-if-changed entries for registry subdirectories.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
src/registry/catalog.rs	Introduces is_registry() and applies it in find_dir(); adds tests for the new predicate.
build.rs	Adds directory-level Cargo rerun triggers for registry subdirectories to detect add/remove changes.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The is_registry heuristic will treat any directory named registry that merely contains an empty tools/, channels/, or mcp-servers/ subdirectory as a valid registry. That can still lead to the original failure mode (finding a “valid” registry that contains no manifests) if a user/project happens to have registry/tools present but empty or unrelated. Consider tightening this to require _bundles.json OR at least one *.json manifest file under one of the known subdirectories (e.g., via read_dir + extension check).

Suggested change

	/// i.e. it contains `_bundles.json` or at least one of the known subdirectories
	/// (`tools/`, `channels/`, `mcp-servers/`). This prevents accidentally matching
	/// unrelated directories named "registry" such as `~/.cargo/registry/`.
	fn is_registry(dir: &Path) -> bool {
	if !dir.is_dir() {
	return false;
	}
	dir.join("_bundles.json").is_file()
	|| dir.join("tools").is_dir()
	|| dir.join("channels").is_dir()
	|| dir.join("mcp-servers").is_dir()
	/// i.e. it contains `_bundles.json` or at least one manifest file in one of the
	/// known subdirectories (`tools/`, `channels/`, `mcp-servers/`). This prevents
	/// accidentally matching unrelated directories named "registry" such as
	/// `~/.cargo/registry/`.
	fn has_json_manifest(dir: &Path, subdir: &str) -> bool {
	let manifest_dir = dir.join(subdir);
	if !manifest_dir.is_dir() {
	return false;
	}
	let Ok(entries) = std::fs::read_dir(&manifest_dir) else {
	return false;
	};
	entries.flatten().any(|entry| {
	entry.path().is_file()
	&& entry
	.path()
	.extension()
	.and_then(|ext| ext.to_str())
	.map(|ext| ext.eq_ignore_ascii_case("json"))
	.unwrap_or(false)
	})
	}
	fn is_registry(dir: &Path) -> bool {
	if !dir.is_dir() {
	return false;
	}
	dir.join("_bundles.json").is_file()
	|| Self::has_json_manifest(dir, "tools")
	|| Self::has_json_manifest(dir, "channels")
	|| Self::has_json_manifest(dir, "mcp-servers")

[gagdiez]

I actually do not agree with this. Hitting an empty registry/ folder was highly likely (I did it multiple times), hitting an empty registry/tools/ or registry/channels/ or registry/mcp-servers seams unlikely in the wild and thus, in case of doing so, should hint that there is actually an empty registry

with that being said, I do think that the real solution is to change the registry logic, and place it inside the config folder (e.g. ~/.ironclaw/registry), but that is work for a different PR

because of that, I assume that this fix is "good enough" as of right now, except if another human reviewer provides a different opinion

[gemini-code-assist]

Code Review

This pull request improves the registry directory detection logic by introducing a helper method is_registry that validates directory contents, preventing accidental matches with unrelated directories. It also updates the build script to include additional directory-level watches for registry subdirectories and adds unit tests to verify the new detection logic. I have no further feedback, except for a minor suggestion to convert a regular comment into a documentation comment for consistency.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.