The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the application's source code and Jira issue descriptions to derive functional modules for component synchronization.
Ingestion points: In references/components.md, the agent is instructed to read UI and API routes (e.g., app/**/page.tsx, app/api/**/route.ts) and project Epics or Stories via bun run jira:sync-issues pull to identify product areas.
Capability inventory: The skill can write to the local filesystem (plan files in .context/reports/ and catalogs in .agents/), execute local scripts, and perform Jira mutations through the REST API.
Boundary markers: The skill enforces a "sealed mutation contract" where the agent must author a plan file and wait for explicit user approval before executing any changes with the --apply flag.
Sanitization: No specific sanitization or escaping of the content read from the codebase or Jira issues is mentioned before it is processed by the agent's logic.
[EXTERNAL_DOWNLOADS]: The skill downloads a manifest file and reference catalogs from GitHub during the instance migration workflow.
Source: Fetches jira-required.yaml from raw.githubusercontent.com. The specific URL path is dynamically extracted from scripts/sync-jira-workflows.ts at runtime.
Usage: The manifest is used to compare local configurations against upstream baselines before regenerating Jira catalogs.
[COMMAND_EXECUTION]: The skill utilizes several local scripts and standard system utilities to perform environment audits and synchronization tasks.
Scripts: Executes local TypeScript files using bun, such as scripts/sync-jira-components.ts, scripts/sync-jira-fields.ts, scripts/sync-link-types.ts, and scripts/sync-workflows.ts.
System Utilities: Uses grep, jq, curl, and ps for tasks like searching the codebase for hardcoded IDs, verifying reachable URLs, and inspecting the process environment for stale variables.

jira-administration