aicb-roslyn-mcp

Overview Schema Source Related Servers Discussions Admin

Prepare Context for an Edit

prepare_task

Read-onlyIdempotent

Build an edit-ready context bundle for a task goal. It covers symbol sources, dependencies, covering tests, and convention siblings, all within a token budget.

Instructions

Build an EDIT-ready context bundle for a task goal: the source of the symbols the goal names, their dependency neighborhood, the test methods that cover them and 1-2 naming/file-convention siblings (the test, the factory, the validator you'd otherwise miss) - as one token-budgeted AI-Builder-MD document.

Prefer this over pack_for_task when you're about to EDIT, not just read. pack_for_task packs the goal-seeded neighborhood alone, without covering tests and siblings. For one symbol use get_context or explain_symbol; for blast radius and test coverage as numbers use review_context.

Returns text, not JSON. It starts with a manifest comment (goal | seeds | tests | siblings | render) so you can see what was covered - and what was NOT ((none)); after a source edit a staleness comment precedes it. 'render' names the path that produced the body. 'template' is the default, because a plain 'aicb mcp' resolves the standard config DB: the chosen facet's template (graphs, detail, line numbers, quality) in the active profile's notation - YAML under the built-in profiles. 'lean' is the fallback of a server without a config DB: tag Markdown. Multi-TFM solutions are deduplicated to one logical project per name (NEWEST TFM's instance, matching find_symbol's view).

Read-only. Parameters: a tight budget drops neighborhood types, tests and siblings first. A type the goal names, or the one owner of a method it names, is never dropped (too large, it renders as structure: members and signatures); the owners of a method name several types share can be dropped. A leading note names the drops. lean and includeQualityMetrics act on the lean render only, so not on a default server.

Input Schema

TableJSON Schema

Name	Required	Description
`goal`	Yes	The task description - name the relevant types/methods in it (e.g. 'refactor OrderService.Cancel and its validation'). Once the goal names a symbol exactly, its other words only steer the trimming and seed nothing of their own; a goal naming no symbol is matched by name segments instead ('cancel order' -> CancelOrder).
`lean`	No	Lean slice (default TRUE): on the lean-render path, drop the repeated <AI_CONTEXT_SPEC>/<AI_CONTEXT_META> preamble and omit enabled-but-empty graph sections ('- none'); keeps PATH_LEGEND / COMPRESSION_LEGEND, the manifest and all non-empty sections. Set false for the full AI-Builder-MD document. (Ignored on the profile-aware slot-template render path, which the active profile controls.)
`facet`	No	Optional facet to work under - the task axis of this call: General (default), Exploration, Refactoring, Debugging, Review, Testing, Documentation, Architecture, or Performance. On a server with a config DB (the default: a plain 'aicb mcp' resolves the standard one) the facet's template shapes the render; it always appends a short work-style trailer with suggested next tools. (The former name 'slot' is still accepted.)
`budget`	No	Optional token budget (floored at 8000). Applies to BOTH the lean and the profile-aware template render path: it compacts/drops method bodies AND drops the least-relevant types when the neighborhood overflows. A type the goal names, and the one owner of a method name it names, is never dropped - one whose own source does not fit renders as structure (members and signatures). The owners of a method name several types share are capped like the neighborhood: some are reduced to structure, some dropped. The bundled covering tests/siblings are BUDGET-BOUND - under a tight budget they are dropped to fit (a leading comment names the drops so you can ask for one by name). Overrides any MCP-profile/template token budget. Omitted, the bundle still renders against a ceiling (the MCP profile's token budget, else the facet template's, else a ~10000-token default budget) that bounds the neighborhood.
`sessionId`	Yes	The session_id returned by analyze_solution. Also takes an absolute .sln/.slnx/.slnf path, analyzed on first use.
`includeTests`	No	Include the test methods that cover the seeded symbols (default true).
`includeSiblings`	No	Include 1-2 naming/file-convention siblings of the seeded types (default true).
`includeQualityMetrics`	No	When true, also emit the <QUALITY_HOTSPOTS> section plus complexity=""/ce="" tag attributes for the bundled symbols (lean path). Default false. Set true for a refactor/quality pass.

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedv0.5.46513Oct 4, 2026

changedInput schema / properties / budget / description

Previous value: -"Optional token budget (floored at 8000). Applies to BOTH the lean and the profile-aware template render path: it compacts/drops method bodies AND drops the least-relevant types when the neighborhood overflows. The goal-named seeds are never dropped - one whose own source does not fit renders as structure (members and signatures); the bundled covering tests/siblings are BUDGET-BOUND - under a tight budget they are dropped to fit (a leading comment names the drops so you can ask for one by name). Overrides any MCP-profile/template token budget. Omitted, the bundle still renders against a ceiling (the MCP profile's token budget, else the facet template's, else a ~10000-token default budget) that bounds the neighborhood."New value: +"Optional token budget (floored at 8000). Applies to BOTH the lean and the profile-aware template render path: it compacts/drops method bodies AND drops the least-relevant types when the neighborhood overflows. A type the goal names, and the one owner of a method name it names, is never dropped - one whose own source does not fit renders as structure (members and signatures). The owners of a method name several types share are capped like the neighborhood: some are reduced to structure, some dropped. The bundled covering tests/siblings are BUDGET-BOUND - under a tight budget they are dropped to fit (a leading comment names the drops so you can ask for one by name). Overrides any MCP-profile/template token budget. Omitted, the bundle still renders against a ceiling (the MCP profile's token budget, else the facet template's, else a ~10000-token default budget) that bounds the neighborhood."

changedInput schema / properties / facet / description

Previous value: -"Optional facet to work under - the task axis of this call: General (default), Exploration, Refactoring, Debugging, Review, Testing, Documentation, Architecture, or Performance. On a profile-aware server (aicb mcp --db-path) the facet's template shapes the render; it always appends a short work-style trailer with suggested next tools. (The former name 'slot' is still accepted.)"New value: +"Optional facet to work under - the task axis of this call: General (default), Exploration, Refactoring, Debugging, Review, Testing, Documentation, Architecture, or Performance. On a server with a config DB (the default: a plain 'aicb mcp' resolves the standard one) the facet's template shapes the render; it always appends a short work-style trailer with suggested next tools. (The former name 'slot' is still accepted.)"

changedInput schema / properties / sessionId / description

Previous value: -"The session_id returned by analyze_solution."New value: +"The session_id returned by analyze_solution. Also takes an absolute .sln/.slnx/.slnf path, analyzed on first use."

First observedOct 2, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Annotations already declare read-only/idempotent/non-destructive, but the description goes well beyond them: return format is text not JSON, the leading manifest comment reveals coverage and gaps ((none)), staleness precedes it after a source edit, multi-TFM solutions dedupe to the newest TFM, and tight budgets drop neighborhood/tests/siblings with a leading note naming the drops. It also documents the never-dropped guarantees for named types and single method owners.

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

Conciseness3/5

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

Front-loaded correctly, with purpose first and usage routing second. However it is long, and substantial portions (the template vs lean render path, facet trailer, budget behavior) duplicate the parameter descriptions already present in the schema, which is padding for a reader who has the schema.

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

Completeness5/5

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

There is no output schema, so the description must carry the return-value burden and does: it explains the manifest comment, staleness line, render/template/lean distinctions, and the text-vs-JSON format. For an 8-parameter, complex bundle-builder, nothing an agent needs to call it correctly is missing.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description adds real meaning beyond the schema: the precise drop order under a tight budget, the never-drop rules, the fact that lean/includeQualityMetrics act only on the lean render path, and the interaction between budget and the profile/template ceiling. These cross-parameter behaviors are not in the schema.

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

Purpose5/5

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

Opens with a specific verb+resource: 'Build an EDIT-ready context bundle for a task goal', then enumerates exactly what the bundle contains (symbol source, dependency neighborhood, covering tests, convention siblings). It actively distinguishes itself from siblings by naming pack_for_task, get_context, explain_symbol, and review_context with the conditions that separate them.

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

Usage Guidelines5/5

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

Explicit routing: 'Prefer this over pack_for_task when you're about to EDIT, not just read', 'For one symbol use get_context or explain_symbol', 'for blast radius and test coverage as numbers use review_context'. Both when-to-use and when-not-to-use are stated, with alternatives named.

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

Deploy Server

Prepare Context for an Edit

Instructions

Input Schema

Schema Changelog

TDQS

Other Tools