Answer: Match the level of instruction specificity to the task's fragility: use high freedom (text-based instructions) when multiple approaches are valid or decisions depend heavily on context, medium freedom (pseudocode or scripts with parameters) when a preferred pattern exists but some variation is acceptable, and low freedom (specific scripts, few parameters) when operations are fragile and error-prone or a specific sequence MUST be followed. Status: Confirmed Entry: tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom Revision: 1 Digest: sha256:217db46533181d3b (changes whenever the content changes) Timeline: Created: 2026-11-13T03:17:45Z Last changed: 2026-11-13T03:17:45Z Last verified: 2026-11-20T12:51:27Z Next check: 2026-11-21T12:51:27Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of High freedom | text-based instructions | [1] | 2026-11-20 Medium freedom | pseudocode or scripts with parameters | [1] | 2026-11-20 Low freedom | specific scripts, few parameters | [1] | 2026-11-20 Guiding principle | Match the level of instruction specificity to the task's fragility. | [1] | 2026-11-20 Sources: [1] Google. "Agent Skill best practices". https://geminicli.com/docs/cli/skills-best-practices.md Role: primary. Accessed: 2026-11-20T12:51:27Z Quote: "**High freedom (text-based instructions)**: Use when multiple approaches are" (Supports: High freedom) Quote: "**Medium freedom (pseudocode or scripts with parameters)**: Use when a" (Supports: Medium freedom) Quote: "**Low freedom (specific scripts, few parameters)**: Use when operations are" (Supports: Low freedom) Quote: "Match the level of instruction specificity to the task's fragility." (Supports: Guiding principle) Published: 2026-10-16T02:36:07Z Snapshot SHA-256: 612c3b8396fdc6669702127975c5d4ec5ce59bd8355fdbfb837fdd83d3d524c3 (captured 2026-11-13T03:17:45Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-13T03:17:45Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T12:51:27Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:51:27Z Last attempt: 2026-11-20T12:51:27Z Notes and limitations: Extracted automatically from a single official source (section "Degrees of freedom"). Related: https://latestn.com/tool/gemini-cli/agents-and-automation/skill-discovery-tiers How does Gemini CLI discover Agent Skills and what are the discovery tiers? https://latestn.com/tool/gemini-cli/agents-and-automation/steer-research-phase How do I steer Gemini CLI's research phase during Plan Mode? https://latestn.com/tool/gemini-cli/agents-and-automation/streaming-json-event-types What event types does Gemini CLI streaming JSON output emit? https://latestn.com/tool/gemini-cli/agents-and-automation/pull-request-ci What CI checks run when I open a pull request in Gemini CLI? https://latestn.com/tool/gemini-cli/agents-and-automation/review-iterate-plan How do I review and iterate on a plan Gemini CLI generated? https://latestn.com/tool/gemini-cli/agents-and-automation/run-headless-mode-single-prompt How do I run Gemini CLI in headless mode with a single prompt? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:17:45Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom POST: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I choose the right level of instruction detail for a Gemini CLI Agent Skill? Revision 1, last verified 2026-11-20T12:51:27Z. https://latestn.com/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom (accessed 2026-11-21T10:05:33Z). @misc{LatestN_tool_gemini_cli_agents_and_automation_skill_degrees_of_freedom_r1, title = {How do I choose the right level of instruction detail for a Gemini CLI Agent Skill?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom}}, note = {Revision 1; digest sha256:217db46533181d3b; last verified 2026-11-20T12:51:27Z; accessed 2026-11-21T10:05:33Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/gemini-cli/agents-and-automation/skill-degrees-of-freedom