The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local CLI and shell commands to manage the archival process.
Runs openspec list --json and openspec status to gather metadata about active changes.
Executes mkdir -p and mv via the shell in Step 8 to relocate change directories into an archive folder.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted data from the local repository files.
Ingestion points: The skill reads file content from openspec/changes/*/tasks.md, openspec/changes/*/specs/ (Step 3), and performs a codebase-wide search for implementation evidence (Step 5) in SKILL.md.
Boundary markers: No specific boundary markers, delimiters, or "ignore instructions" warnings are used when processing the contents of these files.
Capability inventory: The agent can execute file system modifications (mkdir, mv) and local CLI commands (openspec) based on the logic derived from analyzed files, as seen in Steps 8 and 1 of SKILL.md.
Sanitization: No content validation, sanitization, or escaping is performed on the data ingested from the repository before the agent makes decisions or interpolates it into its workflow.

openspec-bulk-archive-change