Semantically search the whole indexed project context. Use this first to find which files are relevant to your task, then pass their paths to load_context_files — do not rely on this tool's snippets alone.

Semantically search only the architecture decision records under .context/decisions/. Use this to find ADRs relevant to your current task, then pass their paths to load_context_files — do not rely on this tool's snippets alone. If you need to search across all files in the project, use search_project_files instead.

Semantically search only past session summaries under .context/sessions/. Use this to find prior session notes relevant to a topic, then pass their paths to load_context_files — do not rely on this tool's snippets alone.

Deterministically find the most recent .context/sessions/*.md file (sorted by filename, not semantic relevance). Pass the returned path to load_context_files to load it — do not rely on this tool's snippets alone.

Load specific .context/-relative files into the active context. Each loaded file is tagged with its path and a SHA-512 hash of its contents so reload_active_context_file can later detect changes. Only pass files you actually need — do not load the whole .context/ tree.

Check whether files currently held in active context (previously loaded via load_context_files) have changed on disk, by comparing their known SHA-512 hash against the current one. Returns fresh tagged content for changed files, a short 'no change' message for unchanged files, and 'not found' for deleted files.

Save a summary of the current session to .context/sessions/YYYY-MM-DD.md. Call this at the end of a session with a concise summary of what was done.

Re-index the .context/ directory into the vector store. Run this after updating project.md, adding ADRs, or refreshing BUNDLE.md.

List repositories accessible via the configured repository provider. In multi-tenant deployments, use this to discover which repositories are available before calling other tools. Optionally filter by organisation name.

List every ADR in .context/decisions/ as a lightweight table (number, title, status, filename). Use this before read_adr/read_adr_section to find which ADR you need, without loading full ADR content.

Read one ADR's full raw content by number or filename, tagged with its path and SHA-512 hash so reload_active_context_file can later detect changes. Prefer read_adr_section when you only need one section.

Read one ADR's title and parsed Status without loading its full content. Returns an explicit message for ADRs using the legacy **Status:** inline format.

List one ADR's top-level (##) section names, in document order.

Read a single named top-level (##) section of one ADR (e.g. 'Context', 'Decision', 'Consequences'). Use list_adr_sections first if you don't know the exact section name.

Semantically search within a single resolved ADR, scoped by number or filename. Use this to find relevant sections/passages inside one ADR you've already identified (e.g. via list_adrs or search_adr_index).

Create a new ADR: allocates the next sequential number, writes a 'Proposed'-status stub with the given title and context, and placeholder text in the remaining sections (Decision, Consequences, Alternatives Considered, ADR Review Discussion). No lock/retry against concurrent creation.

Replace a single named top-level (##) section of one ADR. Rejects the 'Status' section — use update_adr_status for status transitions instead.

Transition one ADR's Status, with lifecycle guardrails: rejects unknown statuses; requires the target ADR to already exist for 'Superseded by ADR-XXXXX'; requires an 'explanation' for unusual (non-adjacent-forward) transitions; requires a populated Decision section (or an 'explanation' to fold into it) before moving to 'Accepted'; and removes the 'ADR Review Discussion' section once 'Accepted' is reached.

Overwrite the full content of .context/project.md.

Replace a single named top-level (##) section of .context/project.md.

Get the interview question set for a bootstrap artifact (currently 'project', for .context/project.md). Ask the user each question, then pass the answers as project_sections to bootstrap_context. Pass project_path to also merge in any repo-supplied custom questions from a .project-bootstrap-questions.yaml file at the repository root.

Atomically scaffold a brand-new .context/ directory for a project that doesn't have one yet: creates decisions/ and sessions/, writes the bundled ADR_CREATE_AND_MANAGEMENT.md and PLANNING_LOOP.md governance docs, writes an interview-driven project.md (answers from get_bootstrap_questions), and creates a governance ADR-00001 establishing the project's ADR process. Every step is additive and idempotent — artifacts that already exist are skipped, not overwritten. If the repository has a .project-bootstrap-questions.yaml file at its root, its questions are merged into the interview set used for project.md.