Zeige einem KI-Code-Assistenten nur denCode, der für die Änderung, die er vornimmt, wirklich zählt.
DiffContext ist ein Kontext-Kcompiler für LLM-Code-Agenten. Gib es ein Python-Repositorium und eine Änderung — einen Git-Diff, einen Branch oder eine einzelnen Funkionsnamen — und es liefert die kleine Menge an Funkionen, die das Modell tatsächlich für diese Änderung sicher braucht: de Aufrufer, die dadurgebrach werden, die Unrterklassen, die eine schreiben, die Teksts, die sie abdecken. Es passt sie in alles Tokenbudget, deinescent; du, und es sagt dem Modell, was es lässt.
Es istebute für Menschen um mit LLMs in reale Cbedase — Agent-Loops, PR-Reviw-Bots, CI-Chechs — " überall, wo du entscheid een musst, was in den Prompt gehört, und das Repositorium zu groß ist, um es zu senden.
Und es bewertet sich selbst: Weist es auf dein Repo, und es grebet deine Git-istorie auf, führt Retrieval gegen echte Co-Change-Paare aus und gibt NULL RESULT aus, wen es nicht passt — das herauszufinden ist das Featur.
Wenn du einen Asistente biittest, eine Funkion in einem 50,000-Zeilen-Projekt zu änderung, hast du drei chlecte Optionen: das gesame Repositorium einzufügen (es passt nicht, und Modelle in sehr großen Kontextschlechter werden), nur diese eine Funkion einzufügen (das Modell zerstört drei Aufrufer, die es nie geshenn hat), oder das Gedoren mit grep zu durchsuchten (grep nicht die Unrterklasse finden kann, die sie überschreibt, oder den Handler, der sie über functools.partial empfängt — wir haben die Trefferquote von grep bei Plateau gemessen, egal wie viel du Budget du gibst).
DiffContext ist die vierte Option. Parse das Repositorium einmalig in einen richigen Abhängigkeig-Graphen, wähle für jede Änderung die wenigen Funktionen aus, die ög & wirklich zahhähen, und packe sie in den [minimalsten] / ref? "einfachsten" prompt.
git change ──► changed functions ──► hybrid retrieval ──► token budget ──► LLM-ready context
graph ∪ BM25 ∪ file top-k + tokensRelated MCP server: Serena
pip install diffcontextKeine r Rä, die Z- Zeit-B bhängigkein, Pyth 3.9+.
Für MCP-Integration (Claude Code / Cursor / Wintrieben):
pip install "diffcontext[mcp]"Siedeoile in docs/MCP.md für die Serverkonfiguration.
Von Quelle r Entwickluring:
git clone https://github.com/trakshan-mishra/Diffcontext.git
cd Diffcontext && pip install -e .diffcontext index /path/to/project # cold: seconds; warm: ~0.02s
diffcontext compile --ref HEAD~1 --max-tokens 8000
diffcontext verify --from-history 20 --calibrateMehr Befehle: UAGE.md. Produktionsrezepte: docs/USE_CASES.md.
diffcontext verifer --from-hisory 20 --calibrate gewängt Tekstfälle aus der deiner Repo-Git-historie und mitcht Retrieval dagegen — und gibt NULL RESULT aus, statt einer deorient-iv "Zah" eine dekorativen Zahle, wenn das Tool scheitert — das herauszufinden ist das Featur.
Ja — endk-zuerend gemessen, nicht über Proxy. Bei 128 TextBench-Python-Aufgaben uber die eigene Test-Suites jedes Repositoriums (gaeohne-LLM-as-Judges), den Kontext versichert pass@1 faaktish ungefär 4x: 5.5% → 25.8%, exaktes Mc-Nemar p < 0.0001.
Zwei Einschränkungen, eigen in benchmarks/context/ESULTS.md §6: (a) die Startfunktionen aller Testbedingungen sind orale — aus dem Gold-Patch extraiert —, das misst also "wenn die Lokalisierung stimmt, zählt dann die Kontextualit?", nicht die "as-solution" des Probles; (b) für 121 der 128 Test[aufgaben] sind dankinet, "as es größtente "Django-Erge" ist.
Die eehrliche "Begleitungen: die drei Kontextvarianten (default / gap / depboost) sind statistisch nicht zu unterscheiden one, p = 0.360–0.81. Der Gewinner ist Kontext versus keinen Kontext — ncht dieser Selektor versus jenerlow. Volständige Ergebnise: Benchmarks/Context/ESULTS.md.
Kein Codegenerator. Es selektiert und packt Kontext; der Code schreibt das Modell.
Nicht präzisions-first. Es wirft ein nettes Netz — die Trefferquote ist unter 0.1 " be default [Top-k . Einsatz --cutoff gap wen du pro Ben nωb bezahlen musst.
Noch nicht mehrsprachig. Pyth- ist voll unterstützt. TypeScript/JS (ESM) ein arbeiten-Prototyp; CommonJS werein gemessen "Fehlermode.
**Kein Ersatz für "Code" lesen. *die statische Anallyse hat bfinde Flecken, genannten luie unten in docs/BENCHES.md.
Grundwahrheit wird aus der Git-historie erzeugt — ein Entwickler änderte diese Funktionen zusammen in einem Commit; zeige eine, ndet das Tool die anderln? Gemessen mit 701 echten Commits in 9 Pythten und 6- (akt), sowie als CI-Gate bei jedem Push" ausgeführ, so dass die Qualität nichtler unbemerkt sinkt.
Per-Commmit "Hit / Recall" für echten Co-Change-Partner, hybrid Retrieal:
django | click | flask | httpx | pydantic | bhack* | r.ack* | |
Hit | 0.894 | 0.889 | 0.863 | 0.935 | 0.78* | 0.897 | 07. `.953 |
Recall | 0.744 | 0.744 | 0.694 | 0.772 | 0.3536 | 0.712 | 0.762 |
* ValidierungsRepos, nie für das Abstimmen verwendet. Die volle Tabelle aller 9 Repos: listet in Benchmars/READ.md.
Mit gleicher Tokenbudget im Griff grep bis 0.215 Recall bei 4k "stagniert", "ährend DiffContext 0.576 at 8k "ensch (2.7×). As der andrte Seite "gehört: die mittlere Präzision ist unter 0.1 beim Standard-"top-k — die meisten abgerufene Symbole sind zweckdienlich Kontext, nicht das exakte "e" Co-Change-S at. The --cutoff gap schneidet die gröesste Score-Differenten-Cut für ~ 4× Präzision und ca. 30% "Recall-Kosten auf " " (Co-Change-Benchmark; 2.2× / ~14% für ContextBgreen).
Ein r 2026-07 "Pass" hat die lauf angegriffen, nicht das To. Drei veröffentichte Zahle "überlebten" nicht:
Kalibrierung — die einzige zitatiole Zahle zuvor (r=0.274, n≈25) wurde auf einem vermutlich verrufenen Index "gemessen. Bei eie "dätige" ine "Berechnung an n=10,8 0.0 und "jetzige Legacy" on "schnapp" den r=old "0.016 (p=0.6)" — "keine Beziehung. Isthliche: Richtung "nicht weiß" → r=0.287 (p=0.0001) — ein Ranking-Signal, keine "wahrscheinlichkeit.
MixedGewichte — die früher [0.5, .0.35, .0.15] warenietée one- Repo-r-bewährt: Jede "Fold" ein anderes, "weniger-graphen" ontententes Gewichtch. Jetzt: [0.3, 0.5, 0.2].
Dens-Base-L-ine — eine TF-IDF-Verer "hätzte RENance "reie" (0.664, 5/5). Die echte MiniLM-Encoder? "0.597* g und "schlägt BM25 nur 2/5" aus. Zwei frühere "Schlussfolgerungen" are "on "recordiertenans" "ans"are "graph".
Volle Abbildung: docs/autiting-my-own-benchmark.md · "Raw-Pass": benchmarks/RIO_Report_2026-07.md führt.
from diffcontext.pipeline import index_repository, analyze_impact, compile
idx = index_repository("/path/to/repo")
impact = analyze_impact(idx, ["./src/auth.py:validate_jwt"])
ctx = compile(idx, impact, max_tokens=8000, top_k=20)
print(ctx.text) # paste-ready, meta-header discloses what was droppedDie inkrementelle API (uff index.update([...]), die strukturierte an "gabe" ( "modeloutput), die Tokenizer.
IdSprache | Status | Qualität der Retrieval |
Pyth-on | Voll | Benchmarikt: 401 Commits, 5 Repos + 4 Validierungs "Repos" |
TypScript / JS- (ESM) | kProtoyp | Mittlere Recall-n 0–68% je nach Coden Stiel |
Skript (CommonJS) | Nicht unterstütz | Gemessen 0.0%% on express — ich empfehle"ml** |
Statische Analyze hat "eieneDecke": Thematische "Geschwister" zwischen keien Aufrufen, und "lbergreifen de** Konseptuelle Verknüpfungen (die "Score er-/off" ist 0/20 for "most" properties), sowie (dyname" "Split": und "bei "zweifelhafter" . Use:
grep -rn "funcion_name(" --include="*.py" . – "before du "do not" on "full " für "no Calers found" "trust ".
docs/ARCHITECTURE.md — Pipeline, "ModulMap", Agent-API
docs/BENCHMARKS.md — alle Zahlen, "Downstream" pass, \
docs/GCP.md — MCP Server for Claude Code / Kursor / Windsurf
docs/ROADMAP.md — means "pr " stark" , "gemessen" "Motivation" with
diffcontext-Service/ – FastAPI-Service + Web-UI
observability/ – "Retrieval-Pipelining-Tracing"13ED7CF9* CONTRIBUTING.md – Einrichtung, CI-Tore, "Adapter-Dev"
MIT
Compile LLM-ready context for a change.
Give it changed symbol IDs (e.g. ./src/auth.py:validate_jwt) or a git ref (e.g. HEAD~1), and it returns the callers, callees, and related functions the model needs to make the change safely — packed into max_tokens with a disclosure header showing what was dropped.
Optionally pass task_description (the bug report or issue text) to bias retrieval toward symbols relevant to the described problem — the one signal the graph alone can't provide.
Args: repo_path: Absolute path to the repository. If omitted, uses the --repo from server startup. changed_symbols: List of changed symbol IDs (e.g. ["./src/auth.py:validate_jwt"]). Mutually exclusive with git_ref. git_ref: Git ref to detect changes from (e.g. "HEAD~1"). Mutually exclusive with changed_symbols. When only task_description is given (no changed_symbols or git_ref), defaults to "HEAD". task_description: The bug report or issue text. Biases retrieval toward symbols semantically related to the described problem, not just structurally near the changed symbols. max_tokens: Token budget for the context (default 8000). meta: Disclosure header level: "full" (default), "compact", or "off". The pass@1 effect of meta level is UNMEASURED.
| Name | Required | Description | Default |
|---|---|---|---|
| meta | No | full | |
| git_ref | No | ||
| repo_path | No | ||
| max_tokens | No | ||
| changed_symbols | No | ||
| task_description | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It exposes useful behavioral details: output is 'packed into max_tokens', a disclosure headers shows what was dropped, retrieval can be biased by task_description, and the pass@1 effect of the meta level is explicitly marked 'UNMEASURED'. These are substantive disclosures, not merely rephrased schema information.
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?
The description is front-loaded with the core action and then expands into concrete input formats, defaults, and caveats. It contains no filler, yet it is rather long due to the detailed Args block. Every sentence earns its place, and the most critical behavioral caveat ('UNMEASURED') is included without unnecessary qualification.
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?
For a 6-parameter tool with no annotations and no input-schema descriptions, the description covers all required behavior: input formats, defaults, output composition, token budget, disclosure header, and task bias. The one gap is explicit behavior when all inputs are omitted: it only explains defaulting to HEAD when 'task_description' is given, not what happens when no parameters at all are passed. This is a small but real completeness gap.
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?
Although input-schema content coverage is 0%, the description's Args block thoroughly documents all six parameters with examples, defaults, mutual exclusivity, and behavioral caveats. It explains the exact meaning of changed_symbols and git_ref, the default git_ref when only task_description is passed, and the enum-like behavior of meta. This fully compensates for the sparse schema.
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?
The opening sentence states a specific verb and resource: 'Compile LLM-ready context for a change.' It further clarifies the resource by naming the accepted inputs (changed symbol IDs or git ref) and the returned content (callers, callees, related functions). It does not explicitly differentiate from the sibling tools, but the purpose is clear enough to be usable in isolation.
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?
The description gives strong input-selection guidance: changed_symbols vs git_ref are mutually exclusive, and task_description alone implies HEAD. However, it says nothing about when to use compile_context over its siblings (find_impact, explain_selection, verify_retrieval), so the when-to-use guidance is limited to parameter choices rather than tool-disambiguation.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Explain why symbols were included or dropped from context.
Returns the included symbols (with scores and token costs) and the dropped symbols (scored but cut by the token budget), so an agent can inspect or filter the selection.
Args: repo_path: Absolute path to the repository. If omitted, uses the --repo from server startup. symbol: The changed symbol ID to build context for. max_tokens: Token budget (default 8000).
| Name | Required | Description | Default |
|---|---|---|---|
| symbol | No | ||
| repo_path | No | ||
| max_tokens | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden here. It is transparent about the behavior: included symbols are returned with scores and token costs, and dropped symbols are scored but cut by the token budget. It does not explicitly state side effects or authorization requirements, but its read-only nature is strongly implied.
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?
The description is front-loaded with the core purpose, then the return behavior, then a compact Args block. Every sentence adds value; there is no filler or unrelated detail.
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?
Despite having no annotations, the description covers the essential aspects needed to invoke the tool: what it does, what it returns, and what each parameter means. The only notable gap is the lack of comparative routing guidance against the sibling tools.
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 property descriptions have 0% coverage, so the description must compensate. It does: repo_path is 'Absolute path to the repository' with a fallback to the --repo startup value, symbol is described as 'The changed symbol ID to build context for,' and max_tokens is defined as a 'Token budget' with a default. This adds meaningful semantics beyond the raw schema.
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?
The description clearly states the tool's purpose with a specific verb and resource: 'Explain why symbols were included or dropped from context.' It also details what is returned. It does not explicitly compare itself to siblings such as compile_context or verify_retrieval, but the purpose is specific enough to be distinguished from them.
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?
The description implies usage context: after context selection, an agent can 'inspect or filter the selection.' It also explains parameter defaults such as using the startup repo when repo_path is omitted. However, it does not explicitly state when to use this tool over its siblings or when not to use it.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Find what breaks if you change a symbol.
Returns the blast radius: direct callers, direct callees, and transitive impact. The "what breaks if I change this" query.
Args: repo_path: Absolute path to the repository. If omitted, uses the --repo from server startup. symbol: Symbol ID (e.g. ./src/auth.py:validate_jwt).
| Name | Required | Description | Default |
|---|---|---|---|
| symbol | No | ||
| repo_path | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
There are no annotations, so the description carries the behavioral burden. It clearly discloses the output: direct callers, direct callees, transitive impact, and even describes behavior when repo_path is omitted, using the server startup --repo. It does not explicitly call itself read-only, but the language 'Find... Returns' strongly implies a query-only operation.
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?
The description is short, front-loaded with the core purpose, and each sentence earns its place. The summary 'The "what breaks if I change this" query' is a useful clarifying mental model rather than unnecessary repetition. The two argument descriptions are compact and relevant.
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?
For a fairly simple 2-parameter lookup tool, the description covers the inputs, the core behavior, and the scope of results, and an output schema exists to fill in any return-format uncertainty. It doesn't go quite further to explicitly mention read-only constraints, staleness, or required indices, but there are no annotations and the description is still sufficiently rich for most AI agents.
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?
Input schema has 0% description coverage, but the description fully compensates for both parameters: repo_path is defined as an absolute path with the startup --repo fallback, and symbol is given an exact Symbol ID example (./src/auth.py:validate_jwt). This adds significant, actionable detail beyond the raw schema.
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?
The description clearly states the tool's purpose: finding what breaks if a symbol is changed, and what it returns: direct callers, direct callees, and transitive impact. It is specific enough to be distinguished from generic helpers, but it does not explicitly differentiate from sibling tools like compile_context or verify_retrieval.
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?
The phrase "The 'what breaks if I change this' query" gives a strong, useful usage frame: call this when you are about to change a symbol and want to assess impact. However, there is no explicit statement about when not to use it, no comparison to alternatives, and no exclusions, so the guidance remains implied rather than explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Mine git history and grade retrieval quality on your repo.
Generates test cases from co-change history, runs DiffContext retrieval against them, and reports hit/recall. Prints NULL RESULT when the tool doesn't fit your repo — finding that out IS the feature.
Args: repo_path: Absolute path to the repository. If omitted, uses the --repo from server startup. n: Maximum number of test cases to generate from git history (default 20).
| Name | Required | Description | Default |
|---|---|---|---|
| n | No | ||
| repo_path | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Even though no annotations were provided, the description reveals significant behavioral traits: it generates test cases from co-change history, runs DiffContext retrieval, reports hit/recall, and optionally prints NULL RESULT when the repo doesn't fit. Calling out the NULL RESULT as a feature is genuinely useful and goes beyond typical descriptions.
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?
The definition is short, front-loaded with the main purpose, then narrows to mechanism and parameters. The NULL RESULT sentence is pointed and earns its place. The Args section provides compact parameter context without re-describing what the schema already defaults.
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?
Given that there is an output schema and only two simple parameters, the description is complete: it covers what the tool does, the algorithm, output metrics, edge-case behavior, and all parameter semantics. An agent has enough information to select and invoke this tool correctly.
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 description coverage is 0%, so the description carries all param documentation. It clearly adds meaning by explaining repo_path as an absolute path with a startup fallback to --repo, and describes n as the maximum number of test cases with a default of 20. This covers both parameters meaningfully.
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?
The description opens with a specific verb and resource: 'Mine git history and grade retrieval quality on your repo,' and then defines the mechanism (generate test cases, run retrieval, report hit/recall). It clearly explains what the tool does, but it does not explicitly contrast it with sibling tools like compile_context or find_impact.
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?
It gives clear context: you use this when you want to assess retrieval quality over git history, and the NULL RESULT warning indicates what to expect if the repository isn't a good fit. However, it does not state explicit when-to-use versus known sibling alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.5.4compile_contextexplain_selectionfind_impactverify_retrievalScored across 4 tools
compile_context and find_impact overlap substantially because both return callers/callees for a symbol or change, requiring an agent to read descriptions carefully to pick the right one. Their intended uses are reasonably distinct—compile_context produces token-budgeted LLM context, while find_impact answers 'what breaks'—and explain_selection and verify_retrieval are clearly separate.
All tool names follow the same snake_case verb_noun pattern: compile_context, find_impact, explain_selection, verify_retrieval. There are no mixed conventions or vague names.
Four tools is an appropriately focused set for this domain: a primary retrieval/context tool, an impact-focused variant, an introspection tool, and an evaluation tool. None of the tools feel redundant or extraneous.
The main workflow is covered: generate context, inspect impacts, explain selection decisions, and verify retrieval quality on the repo. There is a minor gap in that there is no tool to directly explore the raw dependency graph outside these retrieval wrappers, but that is a narrow gap for this tool's stated purpose.
Deterministic context layer for your codebase: change impact, blast radius, answers with receipts.
Project memory, semantic code search, and grounded agent context.
Provide your AI coding tools with token-efficient access to up-to-date technical documentation for…
Codebase intelligence for AI agents — dead code, blast radius, ownership.
