sdd-orient
Skill0 starsChanged 37 days agoMITView on GitHub
Use when locating and inspecting a software project's repository before SDD work, checking Git worktree and interrupted-task state, discovering governing instructions and development documents, or handing a scoped project orientation to an orchestrating agent. Read-only; report whether a proposed mutation has the required Git and authority baseline.
In this repository14 other files it gives its agents.
- sdd-conventionsSkill
- sdd-designSkill
- sdd-docsSkill
- sdd-forgeSkill
- sdd-implementSkill
- sdd-integrate-featureSkill
- sdd-manageSkill
- sdd-planSkill
- sdd-reportSkill
- sdd-specifySkill
- sdd-steerSkill
- sdd-tasksSkill
- sdd-tddSkill
- sdd-verifySkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
If the target is outside a usable Git worktree, mark repository mutation blocked.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Orient in a project
Establish a factual, scoped starting point before another SDD skill changes repository files. This skill is a shared, read-only capability. It neither authorizes a mutation nor decides which later workflow to run. Read inspection and handoff for the evidence checks and report contract.
- Establish the user's intended project path and proposed scope, if supplied. Distinguish the project root from its enclosing Git root; do not assume the current directory is the project.
- Inspect root
SDD-MANAGER.md,AI_DISCLOSURE.mdand README links as adoption/disclosure evidence and report missing or conflicting records in the handoff; do not create them during orientation. Identify applicable repository instructions, including root and path-scopedAGENTS.md, referenced policies, and project-designated sources. Inspectdocs/dev/PROJECT.mdfor the project brief; if it contains operating instructions, preserve their applicability and report their scope and any conflict. - Inspect Git, current branch or detached/unborn state, HEAD, worktree status, and changes affecting the intended paths. Suppress optional Git writes with command-scoped
git --no-optional-locksor an equivalent scopedGIT_OPTIONAL_LOCKS=0. Git commands must not change the index, branch, working tree, or configuration. - Inspect adjacent SPEC/PLAN/TASKS (and selected feature) review reports, finding/gate state and reviewed/governing identities; report absent or apparently stale readiness without performing QC or writing reports. Discover present development documents and their focused children, the active feature or change context, relevant code and tests, and declared project commands. When a task list exists, identify the last committed task and compare it with the owning checklist, pending changes, and existing verification evidence. Distinguish completed work awaiting commit from incomplete work; pass the identified state to sdd-implement for continuation.
- Produce the scoped orientation report. Separate observed facts from inferences and unknowns, identify concrete blockers, and state whether the Git prerequisite for a contemplated mutation is met. Re-orient when the project path, target scope, HEAD, instructions, or material worktree state changes.
If the target is outside a usable Git worktree, mark repository mutation blocked. Do not initialize Git, create files, clean worktrees, restore files, run tests with side effects, commit, or invoke a mutating workflow. A conversational or read-only inspection may continue.
For a subagent, the orchestrator supplies the applicable path scope, instructions, relevant document authorities, Git baseline, known dirty paths, and prohibited mutations. A subagent can inspect additional evidence within its scope; the orchestrator rechecks the overall state before coordinating any mutation or commit.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

