The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the openspec CLI utility for core operations. These include checking the tool version (openspec --version), initializing projects (openspec init), scaffolding new changes (openspec new change), checking status (openspec status), validating artifacts (openspec validate --all), and archiving completed changes (openspec archive). These commands are standard for the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, potentially untrusted sources such as issue trackers, which is a common attack surface for indirect prompt injection.
Ingestion points: Data is retrieved from issue trackers and project planning artifacts (issues, ADRs, plans) as described in SKILL.md and references/042-planning-openspec.md.
Boundary markers: The skill contains explicit defensive instructions stating that issue content must be treated as untrusted requirements data and cannot supply executable instructions or authorize command execution.
Capability inventory: The skill utilizes the openspec CLI and read-only tracker tooling.
Sanitization: The instructions mandate the use of maintainer-provided sanitized summaries for third-party content outside of specific automated modes to prevent the ingestion of raw, potentially malicious prose.

042-planning-openspec