j5ik2o-takt-task-builder
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-generated task specification files (
order.md), creating a potential surface for indirect prompt injection. - Ingestion points: The skill reads
order.mdfiles from the.takt/tasks/directory during its validation phase (Step 5) and via thescripts/validate-order-md.shshell script. - Boundary markers: There are no explicit delimiters or instructions to ignore embedded instructions in the processed data.
- Capability inventory: The skill executes shell commands (
bash) and performs filesystem operations based on task content. - Sanitization: The input is checked for structural consistency using string matching (
grep/awk) but does not undergo semantic sanitization. - [COMMAND_EXECUTION]: The skill instructs the agent to run a bundled shell script (
scripts/validate-order-md.sh) to validate file structures. It also suggests executing scripts from a sibling skill (j5ik2o-takt-workflow-builder) for workflow validation. These operations are core to the skill's functionality for developer environments. - [EXTERNAL_DOWNLOADS]: The documentation includes references and links to official project resources on well-known platforms, including GitHub repositories (
github.com/nrslib/takt), npm, and Discord. These references point to legitimate project support infrastructure.
Audit Metadata