manage_project_context

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description adds useful behavioral color ('bounded', 'durable', 'no re-analyzing unchanged sources') and does not contradict annotations (destructiveHint=false aligns with a 'clear' operation). However, it omits richer details like what happens on clear, permissions needed, or failure modes. It adds some value beyond the annotation flags but doesn't substantially elevate context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Three tight sentences, front-loaded with the primary verb set, followed by a one-sentence elaboration on the two main workflow actions, and closed with a crucial security caveat. Zero waste, every clause earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Adequate for a 4-param tool: the description introduces the workflow, the key actions, and a security boundary. Yet it leaves gaps around expected sequencing (e.g., that enrich logically implies a prior capture), what 'status' returns, and how project_id ties the two halves together. No output schema exists, so the description carries that burden, and it comes up slightly short.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the 3 baseline applies. The description does bridge the action enum to the enrich record types (mapping 'enrich' to transcripts/shots/audio/notes) and hints at the source_revision guard with 'without re-analyzing unchanged sources', yet it adds no syntax or format details beyond the schema's existing high-quality parameter descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

Begins with explicit verbs ('Capture, enrich, inspect, or clear') tied to a specific resource ('durable local Premiere project-context index'). Clearly distinguishes this from standard metadata tools by framing it as an index manager with a capture/enrich workflow, separating it from siblings like get_metadata or get_project_info.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Clearly delineates when to use capture vs. enrich ('without re-analyzing unchanged sources') and enumerates accepted enrich record types. Includes an important constraint ('Never include secrets or unrelated customer data'). However, it never explicitly names alternative tools for simpler lookups, relying on implication rather than direct exclusions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.